Lots of folks talk about search and never delve into the content network. In the search network you choose keywords based on which ads will be shown in search results. In contrast with the content network, you choose locations, ie, web-sites and individual pages within those sites, where you would like to see your ad.

The content network is the set of sites, blogs and other web products that networks like Google and Yahoo push to publishers that have agreed to publish ads in exchange for a fee per click through an advertising program.

The owner of the website which publishes the ads chooses one or more locations where they will appear, and receives a commission for each click received. The first thing to understand is that, unlike search engine searches and resulting ads, advertisements in the content network are not geared for keywords, but by the content of websites that they occur.

David Szetela’s new book Customers Now: Profiting From the New Frontier of Content Based Internet Advertising shows exactly how to use the networks, set up ads and optimize the platforms for better ROI. Googles content network and  ContextWeb’s ADSDAQ Exchange is also reviewed but does not have the reach of Yahoo or Googles network.

Most importantly they sent back a note acknowleding my tweet and expressing their appreciation for my suggestion.

Even better, they thought it was such a good suggestions they immeadiately ran it up the flagpole to the engineering and product development group.

In the end this company now continues to hold my admiration for the product and the company as a whole.

Saying thank you costs a company nothing. Customer feedback that can make your product or service even better?

Priceless.

Abstract

A system for ranking geospatial entities is described. In one embodiment, the system comprises an interface for receiving ranking data about a plurality of geospatial entities and an entity ranking module. The module uses a ranking mechanism to generate place ranks for the geospatial entities based on the ranking data. Ranked entity data generated by the entity ranking module is stored in a database. The entity ranking module may be configured to evaluate a plurality of diverse attributes to determine a total score for a geospatial entity. The entity ranking module may be configured to organize ranked entity data into placemark layers.

    GOOGLE / FENWICK
    SILICON VALLEY CENTER
    801 CALIFORNIA ST.
    MOUNTAIN VIEW
    CA
    94041
    US

1. A geographic information system (GIS) comprising information about a plurality of geospatial entities and configured to prioritize the geospatial entities according to a ranking mechanism.

2. The system of claim 1, wherein the ranking mechanism uses data about a meta attribute of a geospatial entity to determine the geospatial entity’s priority.

3. The system of claim 2, wherein the meta attribute comprises one of: quality of information available about the geospatial entity, an attribute of a description of the geospatial entity, and an attribute of a definition of the geospatial entity.

4. The system of claim 2, wherein the meta attribute comprises an indicator of the geospatial entity’s popularity.

5. The system of claim 2, wherein the meta attribute comprises one of: an age attribute, a stature attribute, and an importance attribute.

6. The system of claim 2, wherein the meta attribute comprises a relationship of a geospatial entity to its place in a hierarchy of geospatial entities.

7. The system of claim 1, wherein an entity of the plurality of entities comprises a collection of geospatial objects and wherein the priority of the entity is determined responsive to a characteristic of the collection of geospatial objects.

8. The system of claim 1, wherein an entity of the plurality of entities comprises a geospatial entity defined in an on-line forum and wherein the ranking mechanism uses data generated in the on-line forum to determine the rank of the geospatial entity.

9. The system of claim 1, wherein the ranking mechanism uses data harvested from a website on the internet about a geospatial entity to determine the geospatial entity’s priority.

10. The system of claim 1, wherein the ranking mechanism determines a geospatial entity’s priority from a combination of weighted data from a plurality of meta attributes of the geospatial entity.

11. A computer-implemented method for ranking geospatial entities, the method comprising: receiving geospatial entity data; evaluating attributes of geospatial entities included in the received geospatial entity data; ranking the geospatial entities based on the evaluation; and storing the ranked geospatial entity data.

12. The method of claim 11, wherein the geospatial entity data comprises data generated in a community forum.

13. The method of claim 11, wherein the geospatial entity data comprises data associated with a specific user and further comprising using the ranked geospatial entity data to generate a map for the specific user.

14. The method of claim 11, further comprising selecting geospatial entities for a geographical display based on the rankings of the geospatial entities.

15. The method of claim 11, further comprising providing the ranked geospatial entity data to a map system configured to generate a map that includes ranked geospatial entities and unranked geospatial entities.

16. The method of claim 11, further comprising selecting geospatial entities to include in navigation instructions based on rankings of the geospatial entities.

17. The method of claim 11, further comprising selecting a geospatial entity to associate with an advertising term based on the geospatial entity’s ranking.

18. The method of claim 11, further comprising providing the ranked geospatial entity data to an application for generating a search result based on the ranked geospatial entity data.

19. The method of claim 11, wherein evaluating is performed responsive to user instructions for providing personalized geospatial entity rankings.

20. The method of claim 19, wherein the user instructions comprise a weighting to be applied to an attribute of a geospatial entity identified in the geospatial entity data.

21. A system for ranking geospatial entities, the system comprising: an interface for receiving ranking data about a plurality of geospatial entities; an entity ranking module for generating place ranks for geospatial entities according to a ranking mechanism based on the ranking data; and a database for storing ranked entity data generated by the entity ranking module.

22. The system of claim 21, wherein the interface is configured to provide the ranked entity data to a requesting application.

23. The system of claim 21, wherein the entity ranking module is configured to evaluate a plurality of diverse attributes to determine a total score for a geospatial entity.

24. The system of claim 21, wherein the entity ranking module is configured to organize ranked entity data into placemark layers.

25. The system of claim 24, wherein each placemark layer corresponds to at least one of: a level of detail, a density, an altitude, and an entity category.

26. The system of claim 21, wherein the requesting application is a map server system configured to use the ranked entity data to generate a map including entities selected on the basis of place ranks.

27. The system of claim 26, wherein the entity ranking module is hosted on the map server system.

28. An entity ranking module hosted on a client device, the module for generating rankings for a plurality of geospatial entities and the module comprising: an interface for receiving entity data that defines a plurality of geospatial entities and ranking data that describes the plurality of geospatial entities; and a ranking engine for generating rankings for the geospatial entities, wherein the rankings are used to select which of the geospatial items to include in a map to be displayed on the client device.

29. The module of claim 28, further comprising a memory for storing data about a user of the client device and wherein the ranking engine is configured to apply a ranking mechanism responsive to the user data.

30. The module of claim 29, wherein the user data comprises user preferences about the relative weightings of attributes evaluated by the ranking engine.

31. The module of claim 29, wherein the user data comprises a user defined geospatial entity.

32. The module of claim 29, wherein the user data comprises an indication of a user’s interest in a geospatial entity and wherein the ranking mechanism assigns a rankings premium to the geospatial entity based on the user’s interest.

RELATED APPLICATION DATA

[0001] This application claims the benefit of U.S. Provisional Application No. 60/726,505, filed on Oct. 12, 2005, which is hereby incorporated by reference in its entirety.

BACKGROUND

[0002] 1. Field

[0003] The invention relates to mapping systems, and more particularly, to techniques for prioritizing geographical entities for placement on geographical displays.

[0004] 2. Description of the Related Art

[0005] A geographic information system (GIS) is a system for archiving, retrieving, and manipulating data that has been stored and indexed according to the geographic coordinates of its elements. The system generally can utilize a variety of data types, such as imagery, maps, and tables. Historically, GIS technology has been used for scientific and governmental investigations (e.g., to identify geographical areas adversely impacted by pollution or over-building), resource management (e.g., regional forestry observation), and development planning (e.g., suburban development of under-utilized geographic areas).

[0006] More recently, GIS technology is being integrated into Internet-based mapping applications. Users can annotate digital map locations with placemarks (e.g., designated on the map with an icon or other graphic). Some placemarks allow the user to write a brief description relevant to the location marked by the placemark, while other placemarks allow the user to change the style of icons and/or labels associated with the placemark. However, in many instances, the number of available placemarks is significant.

[0007] What is needed, therefore, are techniques for prioritizing which placemarks (as well as other map entities) to display on a GIS-based map.

SUMMARY

[0008] The above need is met by techniques for generating prioritized entity data described herein.

[0009] In an embodiment, a geographic information system (GIS) comprises information about a plurality of geospatial entities and is configured to prioritize the geospatial entities according to a ranking mechanism. The ranking mechanism uses data about a meta attribute of a geospatial entity to determine the geospatial entity’s priority. The meta attribute may vary in different implementations but in one embodiment comprises the quality of information available about a geospatial entity.

[0010] In another embodiment, a computer-implemented method can be used to rank geospatial entities. The method comprises several steps including receiving geospatial entity data, evaluating attributes of geospatial entities included in the received geospatial entity data, ranking the geospatial entities based on the evaluation, and storing the ranked geospatial entity data.

[0011] Another embodiment of the present invention provides one or more machine-readable mediums (e.g., one or more compact disks, diskettes, servers, memory sticks, or hard drives) encoded with instructions, that when executed by one or more processors, cause the one or more processors to carry out a process for ranking geospatial entities. This process can be, for example, similar to or a variation of the methodologies described herein.

BRIEF DESCRIPTION OF THE DRAWINGS

[0012] FIG. 1 is a block diagram of a GIS system with entity ranking capabilities in accordance with an embodiment of the invention.

[0013] FIG. 2 is a block diagram of the entity ranking module shown in FIG. 1, according to one embodiment of the present invention.

[0014] FIG. 3 illustrates a method for generating and providing prioritized entities in accordance with an embodiment of the invention.

[0015] FIG. 4 illustrates a method for requesting and receiving a map including prioritized placemarks in accordance with an embodiment of the invention.

[0016] FIG. 5 depicts a GIS map that includes prioritized placemarks in accordance with an embodiment of the invention.

[0017] The figures depict various embodiments of the present invention for purposes of illustration only. One skilled in the art will readily recognize from the following description that alternative embodiments of the structures and methods illustrated herein may be employed without departing from the principles of the invention described herein.

DETAILED DESCRIPTION

[0018] A geographic information system (GIS) is disclosed that provides users with a greater degree of flexibility, utility, and information. The system may also be configured as a distributed geographic information system (DGIS). The system employs techniques for prioritizing which placemarks (as well as other map entities) to display on a GIS-based map.

General Overview

[0019] It is common practice in Geographic Information Systems to provide mechanisms to select a subset of available geographic features for display based on any of several criteria. For example, one might load a database of world cities into such a system and then request to see only those cities within the United States, those cities with populations exceeding one million persons, or perhaps those cities meeting both of these criteria. In this last case, only the markers for New York City, Los Angeles, Chicago, Houston, Philadelphia, San Diego, Detroit, and Dallas would be displayed if 1990 population data were being used.

[0020] Further, some interactive Geographic Information Systems support different feature visibility criteria at different viewing distances. For example, a selection criteria might be constructed that would show only cities exceeding one million in population when the view was of the North American continent (viewing distance 1), then include additional cities exceeding 100,000 in population when the viewpoint is lowered to include a single state in the display area (viewing distance 2), and finally modified to include smaller cities as the view lowers within a state or county (viewing distance 3). These and related techniques are known in the GIS field as `selection` and in the computer image generation field as `visibility culling` and `level of detail management`.

[0021] An embodiment of the present invention is configured to determine `which few of many` geospatial items or entities to display on a map according to the relative importance of each entity according to its rank. Such a rank can be generated based on extrinsic factors, such as the popularity of the entity to users (e.g., quantity and/or velocity of geospatial item access), quality of information (e.g., respect for the information source of geospatial item and community stature of geospatial item’s author), and similar meta-data about the geospatial feature. For example, if many users of an interactive GIS system view a particular small city (e.g., Henderson, Nevada), then an embodiment of the present invention would be to give that city a sufficiently significant place rank for display so as to display it along with the major cities exceeding one million in population. In this way users are shown the geospatial entities most likely to be of interest in the area of their visual search within an interactive geospatial information system. The ranking of geospatial items may further be based on location, distance, or other intrinsic attributes of a geospatial feature such as position and altitude (zoom level).

[0022] Although the ranking of geospatial entities is described herein primarily in the context of choosing entities for display to a user on a map, the ranking may be used for other purposes, as will be apparent in light of this disclosure. Examples include selecting which entities should have associated keywords used for determining and displaying relevant advertising; selecting which entities should be suggested as potential origins, destinations, or waypoints in navigation computations; and other uses where an estimation of a most popular or most interesting subset of geospatial entities is desired. In such applications, ranked entity data generated in accordance with an embodiment of the present invention may be supplied to various systems in addition to or instead of a digital mapping system as required by the context.

[0023] Using such entity ranking information, a two or three-dimensional digital map can be generated that includes placemarks that correspond to geospatial entities. In one such embodiment, a map generated by a GIS may include several types of data about geospatial entities. For instance, the map may include geographic features such as the terrain, infrastructure including roads, rail routes and airports, buildings, and/or borders of a landscape. The map may also be annotated with information about government entities and services such as parks and recreational services, federal, state, or local government landmarks, and community services. These and other annotations may be presented in the form of placemarks belonging to one or more categories, including commercial placemarks that represent businesses, travel placemarks including, for instance, historical sights and tourism destinations, user-defined placemarks that have been identified and named by a user for personal or community use, and/or community placemarks that have been voluntarily defined by members of the public in a forum. In one particular embodiment, information presented on a map is organized into collections that comprise layers, such as a terrain layer, road layer, border layer, community placemark layer, etc. Other layers include `current events`, `history`, and `education`, and indicate the organizational taxonomy of the source from which they were taken. A user can interact with a map and turn on or off various layers of information. In an embodiment, a basic or core layer is provided that includes a basic subset of data (for instance, the terrain, major roads, and political borders), and the user can select additional layers to customize the map view. Various third-party content-providers and advertisers can provide individual layers of data that can be overlaid onto such a basic map.

[0024] As will be understood in light of this disclosure, the placemark ranking methods described herein can be used in combination with any conventional, proprietary, and/or emerging techniques to generate a digital map. In the case of a conventional raster map, for instance, the placemarks and other types of map data are used to create a map in a digital format such as .jpeg, .gif, or .png, at a map server and then delivered to a client. Requests to manipulate or interact with the map, are provided from the client to the server, which in turn generates the requested map view. In the case of a tiled raster map, pre-generated, rasterized images or “tiles” that include placemark data are stored at a map server. When the user submits a map query, the rasterized images are provided to the client, where they are used to create the requested map. Additional views based on, for instance, panning, zooming, or tilting the requested map can be generated at the client using the tiles. Vector-based methods can also be used to generate digital maps in accordance with other embodiments of the invention. In one such particular case, map data, including placemark data, is provided to the client by a map server in the form of vector graphic instructions. The instructions are interpreted by an application at the client in real-time to generate a map for the user. As a user interacts with the map, for instance, by including or excluding various layers including geospatial entities, the map can be dynamically updated at the client to include those layers. Likewise, as the user interacts with the map, for instance, by zooming or panning, the map can be dynamically regenerated at the client to include the new map views.

[0025] Geographic information systems (GIS) are referred to throughout the present disclosure. As is known, a GIS may be implemented as a distributed geographic information system (DGIS), in which, for instance, GIS components are distributed across two or more different computers in different physical locations across a network such as the Internet or a corporate enterprise. Reference is also made herein to Google Earth, a GIS-based digital globe that includes various elements such as servers, clients, and other components and features as will be apparent in light of this disclosure. Reference is also made to Google Earth Community, a forum in which placemarks and entities are created, defined, described, and discussed by members of the participating public. Note that “Google Earth” and “Google Earth Community” and the descriptions provided herein may be protected under other forms of intellectual property, and are used for reference purposes only.

System Architecture

[0026] FIG. 1 is a high-level diagram of a digital map system 100 with entity ranking capabilities in accordance with an embodiment of the invention. The system 100 includes a map server system 150 that is communicatively coupled to one or more clients 110 via a network 160. The map server system 150 is coupled to a database 140 of ranked entity data populated (e.g., offline, or in real-time) by a GIS 170. The GIS 170 is equipped with an entity ranking module 120A that applies a ranking mechanism or algorithm to determine the relative ranks of geospatial entities. These entities may be defined within the GIS 170, for instance, based on data provided to the GIS 170 from various external sources 180. The ranked entities are stored in a database 140 and supplied to the map server system 150, which uses the ranked entity data to generate maps for clients 110. In the system 100 shown, client-side entity ranking modules 120B-C are also present for providing additional entity ranking functionality. For instance, in accordance with an alternative embodiment, the map server system 150 may provide the ranked entity data to clients 110 for client-side generation of maps and ranked entity data layers. Principles of heavy-client and light-client functionality equally apply here, as will be apparent in light of this disclosure.

[0027] Other modules may be included in the system, and illustrated modules may be rearranged and functionalities can be distributed. For instance, the GIS 170 can be integrated into the map server system 150. Similarly, the entity ranking module 120A of the GIS may be a standalone module. There may be a single entity ranking module 120A implemented wholly in or with the GIS system 100, without any entity ranking modules 120B-C on the client side 110. In another embodiment, entity ranking is implemented strictly by entity ranking modules 120B-C at clients 110. Other configurations will be apparent in light of this disclosure, and the present invention is not intended to be limited to any particular one. In this example, the term “module” refers to computer program logic or software for providing the specified functionality. When utilized by a client device 120 or map server system 150, a module may be loaded into memory and executed on a processor. In other embodiments, a module can be implemented in hardware (e.g., gate-level logic), firmware (e.g., a microcontroller with embedded routines for carrying out entity ranking as discussed herein), software, or some combination of hardware, firmware, and/or software.

[0028] The client 110 can be any device that allows a user to access the map server system 150 via the network 160. The client 110 may a device or system configured for computing, such as a personal computer or laptop, a mobile phone, a personal digital assistant, a smartphone, a navigation system located in a vehicle, or a handheld GPS system. Other clients 110 (not shown) may also be in communication with the map server system 150 via the network 160.

[0029] Each client 110 includes an application such as a browser that allows the user to interface and communicate with systems such as the map server system 150 on the network 160, as typically done. Examples of browsers include Microsoft’s Internet Explorer browser, Netscape’s Navigator browser, Mozilla’s Firefox browser, PalmSource’s Web Browser, or any other browsing or application software capable of communicating with network 160. Alternatively or in addition, the client 110 may include an application implemented outside of a browser, such as a specialized mapping or geographic application, from which data on the map server system 150 can be accessed. Interactions with the map server system 150 may be accomplished through a plug-in or other executable architecture implemented locally.

[0030] The GIS 170 can be configured with conventional technology, but further includes an entity ranking module 120A configured in accordance with the principles of the present invention. The GIS 170 receives data from various sources 180 upon which ranked entity data can be determined by the entity ranking module 120A. Both geospatial entities and ranking data by which the geospatial entities can be ranked are represented in the data. These types of data can be provided to the GIS 170 in structured and unstructured form. For instance, while entity data in the form of city names and geographies may be provided in a structured form, ranking data in the form of community comments or ratings, for instance, may be provided in unstructured form. Or, entity data and ranking data may be provided from the same structured source that, for example, identifies a city and its population, or an unstructured source such as a community bulletin board in which an entity is defined and data by which it can be ranked is provided.

[0031] The entity ranking capability of the map system 100 is provided by one or more entity ranking modules 120. The entity ranking module 120 collects entity data and ranking data with which the geospatial entities can be rated. This data may be provided from various sources including the GIS 170, external sources 180, and the client 110. These sources are described in further detail with reference to FIG. 2. The module 120 evaluates the geospatial entities using the ranking data in order to determine a score or rank for each of the entities. In an embodiment, the module 120 also associates the entities with placemarks for a map, and generates groupings or layers of placemarks based on, for instance, a certain placemark density or map view altitude. The resulting entities and entity layers can be stored in a ranked entity database 140, at a client 110, or elsewhere. Each of these capabilities will be described in further detail with reference to FIG. 2.

[0032] In a system that includes server 120A and client-side entity ranking modules 120B-C, the client-side modules 120B-C may provide complementary rankings for use in generating a map for a client 110. In one such embodiment, a server-side entity ranking module 120A provides general placemarks whose rank is determined by a set of general ranking data, while a client-side entity ranking module 120B-C provides personal placemarks that have been ranked using personal data about a user, their behavior, or their preferences.

[0033] The network 160 may be any type of communications network, such as a local area network (e.g., intranet), wide area network (e.g., internet), or some combination thereof. Alternatively, the network 160 may be a direct connection between the client 110 and the map server system 150. In general, the client 110, network 160, and/or map server system 150 may be in communication via any type of wired or wireless connection, using a wide variety of communication protocols.

[0034] The map server system 150 can be implemented with conventional or custom technology. Numerous known server architecture and functionalities can be used to implement a GIS server system. Further, the map server system 150 may include one or more servers operating under a load balancing scheme, with each server (or a combination of servers) configured to respond to and interact with clients 110 via the network 160. In one particular embodiment, the server system 150 is implemented as discussed in U.S. application Ser. No. 10/270,272, filed Oct. 10, 2002, titled “Server for Geospatially Organized Flat File Data,” which is incorporated herein.

[0035] In general, when a user of a client computer 110 enters a search query (e.g., via browser and client side agent), it is put into a request by the client 110, and sent to the map server system 150 via the network 160. The server system 160 then determines what the search query is for, and responds with appropriate data from various sub-systems, such as geo-coders, routing engines, and local search indexes, in a format that the requesting client can use to present the data to the user (e.g., via a browser or other application).

[0036] Used in conjunction with the server system 150, the GIS 170 and ranked entity database 140 provide a map system 100 that serves map and GIS data over the Internet or other network 160. The map system 100 allows users to visualize, select, and explore geographic information (e.g., all over the world or in a particular region). The entity ranking module 120A can be configured to place rank available map data items based on various attributes associated with each geospatial feature (or a subdsdscset of geospatial features). These attributes may be extrinsic or instrinsic attributes of a geospatial feature, represent meta attributes of the feature, and/or reflect the personal behavior of a user. Based on ranking entities according to these attributes, users are shown the geospatial entities most likely to be of interest in the area of their visual search within the interactive GIS.

Entity Ranking Module

[0037] FIG. 2 is a high-level block diagram of an entity ranking module 120 that could be implemented at a client computer or server-side GIS according to one embodiment of the present invention. The entity ranking module 120 receives or collects data from various sources about geospatial entities through an interface 250. The data contains both entity data 210 that identifies and defines geospatial entities and ranking data 220 by which geospatial entities can be evaluated. The entity data 210 may comprise placemark-level data. The ranking engine 230 applies one or more ranking algorithms or rank mechanisms to the ranking data 220 to determine scores for geospatial entities defined in the entity data 210. The resulting ranked entity data may be provided through the interface 250 to a requesting application such as a mapping application. Or, the ranked entity data can be formed into placemark layers by a placemark layer generator 240. In whatever form, the ranked entity data is stored and provided to a map server system or other application. The entity ranking module 120 includes a memory 260 in which data collected, including from a client, can be stored.

[0038] The entity ranking module 120 can receive entity data 210 and ranking data 220 about entities from any number of sources. The data may include satellite data, aerial photographs, street-level photographs, digital map data, tabular data (e.g., digital yellow and white pages), and targeted database data (e.g., databases of diners, restaurants, museums, and/or schools; databases of seismic activity; database of national monuments; etc). It may also include government census and population data, building plan data, demographic data including socio-economic attributes associated with a geospatial entity such as a zip code or town, and alternative name data. In one particular embodiment, the data comprises proprietary content collected by a third-party provider and placemarks derived from it can only be accessed by users who have specifically paid for or subscribed to it.

[0039] While these sources comprise structured data about geospatial entities, definitions of geospatial entities 210 and ranking data 220 in the form of information about attributes of geospatial entities may also be provided in unstructured form. Such data can be harvested from websites on the internet, and/or culled or provided from various sources including community forums such as the Google Earth Community, online bulletin boards, or other virtual spaces in which geospatial entities may be defined and described by users in a public, private, or semi-public setting. In the case of the Google Earth Community, for example, an entity may be posted by a user, and then descriptions of the entities may be provided on subsequent postings or replies to the initial posting. The entity ranking module 120 may also receive data from one or more clients 110 that may be particular to a user or client device 110. As described in more detail below, this data can be used to customize rankings and/or a user’s experience.

[0040] Example geospatial entities include a city name and location, a user defined entity, a commercial entity, a geospatial item found in a web search, or any item (e.g., physical thing, event, or quality) having a geographic association. A geospatial entity is thus comprised of a geometry associated with a physical place (such as a set of geographic coordinates on Earth or the moon) and a description. In the case of a geospatial entity that is non-geographical in nature, such as the War of 1812, this geometry may correspond to locations associated with the event. Thus, the entity can correspond to single or multiple physical places and descriptions. For instance, geospatial entities in Google Earth can be either singleton objects or may be a hierarchical folder of objects, each object in which may be another folder or an entity. Thus, while some entities represent one geospatial object, other entities may have folders that in aggregate represent many geospatial objects. A single entity, in turn, may correspond to one or more placemarks. For instance, an entity like “Oakland gas stations” may include several different physical locations, each of which is represented by a separate placemark.

[0041] Ranking data 220 may describe attributes of entities that can be evaluated by the ranking engine 230 to determine the entity’s rank. In an embodiment, the attribute defines the interestingness of an entity to a particular user. Such interestingness can be used to rank the various geospatial entities in the area of a user’s visual search within an interactive geospatial information system (such as Google Earth), so that client-side entity display prioritizing is enabled. As will be explained in turn, “interestingness” for a geospatial entity can be determined by measuring or otherwise determining various types of extrinsic data associated with that entity. In one such embodiment, this measure, scaled by a corresponding weight, forms a bonus that augments the entity’s score or rank (e.g., by addition or multiplication). Thus, higher ranked entities can be given priority for display over lower ranked (less interesting) entities. Data intrinsic to the GIS system may also be considered, as normally done (e.g., zoom level).

[0042] In another embodiment, ranking data 220 comprises various indications of a user’s interest in certain placemarks. For instance, placemarks that have been saved or annotated by the user at the browser or application level could be deemed to be of greater interest to a user. A user’s search terms or patterns of web page access or use may also be correlated to certain geospatial entities and used by an entity ranking module 120 at the client or on a server to select placemarks for the user. In addition, placemarks that the user has defined for his or her own use may be assumed to be of high personal interest. In one such embodiment, geospatial entities including points of interest or personal relevance to the user, such as the location of the user’s house, workplace, child’s daycare, or favorite playground are identified and marked on any map in the vicinity of these elements, regardless of their relative rank as calculated by a GIS. These and other indications of user interest may be gauged from the user’s behavior, or may be in the form of preferences or instructions regarding entities affirmatively provided by the user, for instance instructing the inclusion or exclusion of specific entities or groups of entities in maps provided by a map server system. A rankings premium may be assigned to geospatial entities based on the user’s interest or preferences. User data collected at a client may be stored in the memory 260 of the entity ranking module and used by the ranking engine 230 to generate entity rankings that are personal to the user.

[0043] The ranking engine 230 comprises a module for ranking entities based on descriptions of entity attributes included in the ranking data 220. Depending on the type of data provided, the ranking engine 230 can use a variety of mechanisms to evaluate geospatial entities, which are further described below.

[0044] The entities as ranked by the ranking engine 230 are organized into layers by a placemark layer generator 240. This may be accomplished by determining a level of detail and threshold to be associated, for instance, with a given altitude or density. For instance, when a user’s query implicates a number of entities greater than a given threshold, only those entities having a place rank above a certain threshold are provided. For example, assume the given threshold for the total number of entities that can be displayed at the current map view is 50, and that the place rank threshold is 80. If a user’s query implicates over 100 geospatial entities, and 35 entities have a place rank over 80, then the server system will serve those 35 entities along with the 15 next highest ranked entities for display at the requesting client. Alternatively, all of the implicated geospatial entities generated are served to a client-side entity ranking module 120, which then determines which of those entities to display (in a similar fashion to the server-side functionality). Or, both server-side and client-side entity ranking can be carried out, where the server system serves a set of ranked entities, and a client then displays a sub-set of that served set. Placemarks can be subdivided into layers according to criteria other than altitude or density, including conceptual, spatial, temporal, or other groupings. In one particular embodiment, the placemark layer generator also applies stylings to the various placemarks and stores those with the placemark layers.

Ranking Mechanisms

[0045]

[0046] One embodiment of the present invention is a method for computing a relative ranking of a geospatial entity such as a city name and location, user defined entity, commercial entity, or geospatial item found in a web search, as compared to other such entities. These relative rankings are determined by a ranking engine 230 and are used within a GIS (e.g., as discussed with reference to client-side entity ranking of FIG. 1) to order the display of entities when not all can be selected for display, for example, to show some of the highest or lowest ranked entities, or perhaps a selected range of entities.

[0047] In an embodiment, this ranking, which can be referred to as place rank, is computed based on the weighted contributions of various non-cartographic meta attributes about a geospatial entity. Rather than directly measuring a characteristic of a physical place, such as its population, these attributes reflect traits of abstractions or representations associated with the geospatial entity. Examples include an attribute of a description of an entity (for instance, the amount of detail in the description of an entity or the number of times a description has been viewed), an attribute of a definition of an entity (e.g. the context or downloads of a definition of an entity, or attributes about the creation of an entity in a public forum), an indicator of the popularity of a geospatial entity (such as the number of views, downloads, or clicks on the entity or a placemark associated with the entity or an attribute based on a ranking or score assigned to an entity), or the relationship of an entity to its context, such as the category to which an entity belongs. Attributes that fit into each of these categories are described in greater detail below: [0048] The amount of detail in the description of an entity. Longer descriptions are given a greater score than shorter descriptions, based on the presumption that more words means more information. In one embodiment of the ranking system described herein, this bonus (increased score) is based on the number of characters in the description text irrespective of the number of bytes required to represent that character in a chosen alphabetic encoding (e.g., thereby normalizing for languages that require more data to encode). A detail penalty can be assessed when the description size is less than a selectable threshold as a means to down-weight “empty” or “signature only” texts. [0049] The number of entities created by a particular author or source. Sources that provide many entities are presumed more reliable than sources having created few entities. The rank bonus here rewards effort and experience and applies to user created entities (places or objects defined by users) as well as those taken from larger databases and from commercial vendors. [0050] The context of a publicly posted geospatial entity. Some geospatial entities originate in public forums (such as the Google Earth Community), and through this origination become associated with the context of that forum (e.g., social life in Rome). Since the context of the user’s search is also known, an embodiment of the ranking system described herein gives a rank bonus to entities posted in forums having a common or similar context with the user. [0051] The number of replies to a publicly posted geospatial entity. Some geospatial entities originate in public forums (such as the Google Earth Community) and in this context there is a potential for dialog about such an entity, where some post to a forum is the specification of the entity and subsequent posts are replies to that entity specification post with discussion or additional information. In these cases, an entity can be credited with a rank bonus based on the number of replies. The replies can be selectively counted, either automatically or manually by a moderator, in order to avoid counting non-relevant replies (e.g., off-topic, flames, or the like). [0052] The number of views of a publicly posted geospatial entity. Some geospatial entities originate in public forums (such as the Google Earth Community) and in this context there is a potential for readers of these forums to view some such posts more frequently than others. Since the frequency of viewing (or similarly actions, such as repeat viewing or bookmarking) reflects an interest in the item, an embodiment of the ranking system described herein gives a rank bonus to entities based on the number of views to the web page where the entity is described. When considered temporally, the incidence of entity views may have occurred non-uniformly over time, such as the case with a publicity spike attending an entity associated with a natural disaster. In such cases it may be preferable, for example, to consider 1000 views spread broadly over a sample interval as indicating greater interest, rather than considering 900 views in a short interval followed by 100 views over a longer interval. To reflect this variation in level of interest, an embodiment of the ranking system described herein allows the number of views to be optionally considered as a time series and apply statistical measures to analyze the distribution of interest. In one such embodiment, the entity rank bonus is greater for distributions that are even than those that are uneven, though the choice of a negative value as the weight for this bonus will effectively invert the class of distribution considered most interesting. [0053] The number of downloads of a publicly posted geospatial entity. Some geospatial entities originate in public forums (such as the Google Earth Community) and in this context there is a potential for readers of these forums to download entity data from some such posts more frequently than others. Since the frequency of downloading (e.g., user click-through to the entity) reflects a deep interest in the item, an embodiment of the ranking system described herein gives a rank bonus to entities based on the number times an entity is downloaded. When considered temporally, the incidence of entity downloads may have occurred non-uniformly over time, such as the case with a publicity spike attending an entity associated with a natural disaster. In such cases it may be preferable, for example, to consider 1000 downloads spread broadly over a sample interval as indicating greater interest, rather than considering 900 downloads in a short interval followed by 100 downloads over a longer interval. To reflect this variation in level of interest, an embodiment of the ranking system described herein allows the number of downloads to be optionally considered as a time series and apply statistical measures to analyze the distribution of interest. In one such embodiment, the entity rank bonus is greater for distributions that are even than those that are uneven, though the choice of a negative value as the weight for this bonus will effectively invert the class of distribution considered most interesting. [0054] The community stature of the author a publicly posted geospatial entity. Some geospatial entities originate in public forums (such as the Google Earth Community) and in this context there is an identified author for each entity. In these forums the authors may have a community stature or status based on number of posts, quality of posts as perceived by readers of those forums, common knowledge of an author’s stature, membership level (e.g., moderator, member, paid member, guest, etc.), and similar factors. An embodiment of the ranking system described herein gives a rank bonus to entities based on the stature of its author within such an online community. [0055] The number of entities contained within a multiple entity ensemble. For instance, geospatial entities in Google Earth can be either singleton objects or may be a hierarchical folder of objects, each object in which may be another folder or an entity. Due to this, some entities represent one geospatial object while other entities may have folders that in aggregate represent many geospatial objects. An embodiment of the ranking system described herein gives a rank bonus to aggregated collections since they contain more information. The bonus can be based, for example, on the
total number of locations in the entity (e.g., bonus=5 if 5 locations; bonus=7 if 10 locations; bonus=10 if 15 locations; and bonus=15 if 20 or more locations). A further embodiment of the ranking system described herein assigns a greater rank bonus to multiple entity ensembles when they have more internal folders rather than less internal folders. This reflects the perception that entities with a detailed segmentation using folders are likely to be more detailed in other aspects as well. In practice, this bonus is computed based on the number of folders contained within a multiple entity ensemble, either proportionally to the number of folders or in a segmented manner (e.g., bonus=5 if 5 folders; bonus=7 if 10 folders; bonus=10 if 15 folders; and bonus=15 if 20 or more folders). [0056] The relative age of an entity definition. Some geospatial entities originate in public forums (such as the Google Earth Community) and in this context there is an identified creation date for each entity. When two or more entities are similar in location they likely describe related information. To establish a further refined ranking score for such cases, an embodiment of the ranking system described herein gives a rank bonus to the older post as an encouragement for users to create entities defining new information rather than revisiting previously described locations. In another embodiment, newer data receives a higher rank because it reflects fresher and more current information. [0057] The hierarchical importance of an entity. Some geospatial entities originate in public forums (such as the Google Earth Community) and in this context there is a potential parent-child relationship for the post that defines an entity. A parent post, such as “universities in Kansas” may have several child posts (known as replies) that supply geospatial entities defining each university within the state of Kansas. The parent post, perhaps describing the location of the main administrative office of the Kansas university system, might have an individual score lower than that of its children, which identify schools attended by many thousands of individuals. To direct attention to the parent post in such cases, an embodiment of the ranking system described herein assigns a rank bonus equal to the sum of the scores of any replies to each post. The ranking is thus based on the relationship of a geospatial entity to its place in a hierarchy of geospatial entities. This assures that the initiating parent post’s score is greater than or equal to the sum of its parts. Other aggregation functions may also be used. [0058] The relative importance of an entity category or source of an entity category. Some geospatial entities originate in public forums (such as the Google Earth Community) and in this context there are divisions known as `forums`, within one of which the post that defines an entity will be located. Forums are typically topical; examples include `current events,` `history,` and `education.` The importance of these forums is not necessarily uniform. For example, `current events` entities may be less important over time than `history` entities, which are timeless. This situation is captured in an embodiment of the ranking system described herein by assigning an importance to each source of entities, or categories of entities, for instance, per forum for the Google Earth Community (or other interactive GIS), by vendor for commercial entity data, and similarly for other sources of entity data such as discovery by web search (e.g., low importance) or from official government sources (e.g., high importance). This relative source importance factor scales (e.g., multiplies) the overall score computed for an entity. [0059] The externally-ranked importance of an entity. Some geospatial entities originate in public forums (such as the Google Earth Community) and in this context there may be a score or rank associated with that entity. In the Google Earth Community Forum, for example, each ranked post may have from one to five “stars,” where five stars indicates a well-respected or otherwise highly regarded entity and one star corresponds to a poor or perhaps less respected entity. The metrics used to determine these external rankings are not necessarily uniform across multiple entity sources, so for example the entities from a Google Earth forum might have originating scores ranging from one to five stars, while geospatial entities found at a restaurant review website might be rated with zero to four “forks.” Further, not only are such metrics as “stars” and “forks” not directly compatible in origin and range, the sources of these metrics may have varying degrees of reliability. For example, on source might be a public website with anonymous contributions and ratings while the other might be a professionally-created restaurant or tourist guide reporting refereed evaluations. For these reasons, the method by which an embodiment of the ranking system described herein incorporates such data is prefixed with an initial scale (A) and offset (B) transformation (e.g., NEW=A*OLD+B) before the subsequent weighting process (e.g., AttributeScore=Weight * [NEW]). For example, if a stars-based rating scheme goes from 1 to 5 (i.e, 0% to 100% satisfaction) and a forks-based scheme goes from 0 to 4 (i.e., 0% to 100% satisfaction), then the ratings cannot be compared or processed just by multiplying, since they start at a different values (1 vs. 0). One technique for handling this case of diverse rating systems is to take the entity source’s native range (Low . . . High) and compute the following: Rescaled=(NativeValue-Low)/(High-Low). This maps any range into 0 as low and 1 as high. Then, the result (Rescaled) can be multiplied by a weighting factor (e.g., 100) to allow the entity to gain from 0 to 100 points of rank bonus as the native ranking moves from 1 . . . 5 stars or 0 . . . 4 forks. The values of Low and High are constant for each entity source (Forks, Stars, etc). Using algebra, the resealing equation can be reordered as: Rescaled=[1/(High-Low)] * NativeValue+[Low/(Low-High)] or, A=1/(High-Low) and B=Low/(Low-High). It follows then that Rescaled=A*Native+B. The entity score can then be augmented by WeightForThisAttribute*Rescaled (for each entity obtained from that source). One extension of such an embodiment uses the place ranking computed for an entity by the place rank process described herein to update the external-rank provided at the source where the entity was originally located, which could be for example the Google Earth Community Forums or other GIS forums. [0060] The spatial density of entities in a defined neighborhood. When multiple entities are clustered in a relatively small geographical region, this signifies that authors of the entities have indicated a geographical region of elevated interest. From this it can be assumed that an entity with an elevated density of neighboring entities has a greater value than would otherwise be the case. This is implemented in an embodiment of the ranking system described herein by adding or otherwise providing a rank bonus based on the number of other entities within a defined area that includes an entity’s location (e.g., circle or box centered at an entity’s location). Note that this metric, though geospatial in concept, is not an inherent attribute of an entity but rather an observation of that entity in relation to other entities.

[0061] The click-through rate of displayed entities. When entities are displayed in an interactive GIS such as Google Earth, it is possible to track the number of instances where users further investigate an entity by various means such as clicking on it with a mouse operation, by performing operations on that entity such as selection or highlighting; by using that entity as a routing origin, destination, or waypoint; or by finding that entity to be responsive to a search for information (e.g., as when the entity description contains the word “Pizza” and a user of the interactive system performs a search with the keyword “Pizza”, “Italian”, or “Food”). A tabulation of such events yields a direct measure of importance for that entity as understood by users of the system. When this data is collected (e.g., in the optional client entity ranking module and/or in the server-side entity ranking module) it may be used as an attribute defining interestingness for an entity. An embodiment of the ranking system described herein exploits this measured activity level by assigning a rank bonus to an entity based on the number of user events it receives compared to the average value for other entities. This measure (e.g., scaled by the corresponding weight) forms a bonus added (e.g., or otherwise used) to augment an entity’s score. [0062] The enablement rate of entity categories. When entities are displayed in an interactive GIS such as Google Earth, they are often presented in the form of layers, as described earlier. These layers are selectable individually and in aggregate. The enablement of each layer can be viewed as a vote in a plurality voting system and the aggregation of such votes may be used to determine the relative preferences that users have for the various entity data layers. When this data is collected (e.g., in the optional client entity ranking module and/or in the server-side entity ranking module) it may be used as an attribute defining relative interestingness for a collection of entities. For example, a layer of “Bars and Night Clubs” will likely be enabled more often than a layer of state “IRS Offices”. An embodiment of the ranking system described herein exploits this measured activity level by assigning a relative weighting for a class of entities or source of entities based on the number of layer selection actions observed compared to an average value. This measure can be used to set the relative importance of an entity category as previously discussed, or may be used to scale those static priorities by an additional amount based on empirical observation of user preferences. [0063] The estimated importance of the web page or document associated with an entity. When an entity is linked from a web page or other electronic document, it is possible to use the estimated importance of that page or document to infer the estimated importance of the linked geospatial entity. One estimate of relative importance for online documents is a page rank, such as described in U.S. Pat. No. 6,285,999, titled “Method for Node Ranking in a Linked Database” and/or U.S. Pat. No. 6,799,176, titled “Method for Scoring Documents in a Linked Database” and/or U.S. Patent Application No. 20050071741, titled “Information Retrieval Based on Historical Data,” each of which are hereby incorporated by reference herein. An embodiment of the ranking system described herein uses such page rank scores as a proxy score for geospatial entities referenced by online documents identified in a search result. For example, a Google search for “Kansas university” returns 1,560,000 results, which are ordered by the page rank mechanism described in the ’999 patent to display the ten results deemed most relevant as the initial response to the query. These first ten results are exactly those having the highest page rank score. An embodiment of the present invention uses one or more of search result document page rank scores, scaled by the corresponding weighting factor, as an importance bonus for geospatial entities referenced by those documents. For example, the top page rank score may be used, or the average of the top ten page rank scores, or some other function or selection thereof.

[0064] One or more of the attributes listed above is evaluated and the results are weighted individually with a scale factor before being summed to produce an overall score for an entity. Such a calculation could be performed by the ranking engine depicted in FIG. 2. In an embodiment, both meta attribute data about a geospatial entity and data that reflects attributes of the entity itself are used to determine the overall score. For instance, in an embodiment, the size of a business and the number of replies to a posting about the business are both factors that are weighed by a ranking engine 230. The ordering of entities by non-increasing score defines the ranking (descending order). The weighting factors can be changed based on resulting rankings to adjust the relative prominence of each attribute (e.g., automatically, or as desired by the system administrator, or by an individual user). In one embodiment, an end user can assign the weights to be applied to each type of attribute. Any of the weightings can be set to zero (or one in the case of relative importance) to remove that attribute from consideration or equivalently some of the attributes may be ignored. The weighting factors can be made negative to downgrade rather than reward entities for a particular attribute. Applying scale factors to individual attributes allows diverse attributes to be evaluated and a total score to be calculated based on the attributes.

[0065] Numerous algorithms can be used to for determine place rank in accordance with an embodiment of the present invention. For example, in an embodiment, SCORE=NumberOfPostCharacters+a * NumberOfDescriptionCharacters)+b* NumberOfReplies+c* NumberOfViews+d * NumberOfDownloads+StatureWithinCommunity, wherein a, b, c, and d represent variables that can be adjusted based on the value assigned to each attribute. The StatureWithinCommunity reflects the poster of a description, and, in an embodiment, can range from 200 to 500 depending on the poster’s stature based on any of a number of criteria including reputation, posting behavior, and ratings or endorsements by other users.

[0066] One embodiment of the ranking system described herein aggregates the individual attribute measures with a general linear combination. A more sophisticated aggregation, of which the linear combination is the subset along the main diagonal, is to use an M by M matrix of weights, where the dimension M corresponds to the number of attributes, and matrix reduction is used to allow specification of weights for the full cross product of attributes. For example, such a technique allows a specified weight for the product of `description length` and `author stature` as a combined component. The two equation structures described herein are representative of the variety of attribute combination methods that will be apparent in light of this disclosure, such as including exponentiation or other algebraic forms in the evaluation of each attribute.

[0067] Thus, extrinsic meta data about a geospatial entity can be used as input to a scoring system that produces a relative ranking for purposes such as selection for display within a geospatial information system. There are other attributes that may be used in such a system, such as the degree of match between search terms and entity description text and historical search preferences of an individual user or users in aggregate, that may be incorporated into an interactive GIS as described herein as elements used to determine an entity score. Further, as described earlier, the ranking may be used for purposes other than selection for display.

[0068] FIGS. 3 and 4 are flow charts illustrating steps performed to generate and serve ranked entity data in accordance with embodiments of the invention. Not every possible step is shown in the figures. In addition, some embodiments perform different steps in addition to, or instead of, the ones described herein. Furthermore, the order of the steps can vary from that described herein.

Generating Prioritized Placemarks for a Map

[0069] FIG. 3 illustrates a method for generating and providing prioritized entities in accordance with an embodiment of the invention. One or more steps may be performed by an entity ranking module such as the one depicted in FIG. 2 and implemented on the client and/or server side as shown in FIG. 1. In addition, while certain steps (for instance 350) are triggered by a user request for a map, others may be performed `off-line` outside the context of a particular search request.

[0070] On a regular basis, entity data is received 310 from various sources such as the data sources discussed with reference to FIG. 1. Ranking data is also received 320, from the same or different sources as the entity data. The collection of data may take place, for instance, as part of a database construction process, or a web crawl and indexing process. Such data may also be pushed from vendors or third party sources. Some portions of the data, for instance, data about a user’s personal placemarks, may be updated in real-time in response to a user request while others are received or refreshed according to a regular schedule. Various mechanisms such as those described above may be applied to the ranking data to determine 330 rankings or scores for the entities. Comparable attribute scores based on disparate types of ranking data obtained from various sources can be calculated using the techniques described above. The entity data may be stored in a database such as the one shown in FIG. 1, or in another embodiment, a subset of the ranked entity data is cached in a memory of a client hosting an entity ranking module that determines 330 entity rankings based on the cached data.

[0071] According to the example method shown in FIG. 3, ranked entity data is used to generate 340 layers of placemarks for display on maps. In an embodiment, the entities are mapped to placemarks, and stylings are applied based on the category in which an entity or placemark falls. Groups of placemarks are correlated to levels of detail in order to present a viewer of a map with an appropriate density of placemarks. The levels of detail may reflect any number of inputs, including the altitude at which a map is being viewed, the resolution of the map, other layers depicted on a given map, and user preferences about how sparse or dense a map should be. For instance, in an embodiment, if a placemark score is beyond a certain threshold, it receives a level 5 rating, meaning that map views at or below a certain altitude should include the placemark. The placemarks may also be organized into categories such that an individual placemark can be associated with different groups of placemarks, e.g. placemarks associated with a specific city as well as placemarks of tourist destinations. Each of these categories may be reflected a separate placemark layer.

[0072] In an embodiment, at a later point in time, a request for a placemark layer or layers is received 345, and the appropriate layers are provided 350 to the requester. As discussed earlier, a digital map may be generated according to raster, tiled, or vector-based methods. Depending on the method used, a placemark layer may be requested by a map server in real-time in response to a query, or provided 350 to create map components that are pre-stored and only served when a user request is received. In an embodiment, the placemark layers are provided 350 in response to user preferences or selections. In an embodiment, once provided, the placemark layer is combined with other layers to form a map, at a client or server site.

Requesting a Map with Prioritized Placemarks

[0073] FIG. 4 is a flow chart illustrating steps for requesting and receiving a map including prioritized placemarks in accordance with an embodiment of the invention. FIG. 5. illustrates a map that could be received based on a request formulated using steps in the processes depicted in FIG. 4. One or more of these steps may be performed at a client requesting a map from a map server as shown in FIG. 1. The process begins when a map is requested 410 by a client, for instance from within a browser, specialized application such as Google Earth, or other software application. The map request identifies the geographic area for which the map is sought and the altitude the map should represent. User or client data and preferences are also provided 420 to a map server system. For instance, in an embodiment, the country from which the request originates, which can be determined based on the IP address of the request, for instance, and a user’s language preferences are provided to a map server so that only placemarks in the appropriate language and which reflect country or culture-specific stylings or sources of data are served. The country and language information may be provided 420 from the requesting application or another source such as the operating system of the client.

[0074] In addition, user placemark preferences may also be provided 420 to the map system. The preferences may reflect for instance, which categories or layers of placemarks to display, how many placemark to include (density), and how much identifying information should be displayed on a map. For instance, in the case of the map of FIG. 5, a user can select, using checkboxes on a graphical user interface 520, which layers should be included on the map displayed 550. Each layer of placemarks is associated with its own icon or styling, for instance, a knife and fork icon 51 OA for a dining placemark, and a cup 510B for a drinking establishment.

[0075] In an embodiment, preferences provided 420 to the map system can be used to personalize the selection and display of placemarks to the user. Personalized placemark selection may be accomplished in a variety of ways, for instance by personalizing entity ranking by using a user’s behavior or usage patterns as the basis of an attribute upon which an entity’s rank is scored or varying the weight given to certain attributes based on a user’s input; or, for example, by overriding generalized ranking schemes by always including placemarks that the user has defined or designated herself.

[0076] In the flow chart of FIG. 4, a requester may receive 432 various types of data depending on the personalization preferences provided to the map server system. However, as one of skill in the art would know, other steps reflecting different map generation and customization processes are also possible. In a default embodiment, there are no personalized placemarks 430A, and the client receives 432A and displays 450A a map with general placemarks. In such an embodiment, no entity ranking capabilities are implemented locally at the client. In another embodiment, however, a client requests that the requested map include only personalized placemarks 430C. In such an embodiment, entity ranking capabilities are implemented wholly at the client, with an entity ranking module such as the module described in FIG. 2 hosted locally rather than on a centralized server. The client receives 432A a map, comprised of map layers, which does not include any placemarks on it. Personalized placemarks generated and stored at the client are added 440A to the map, and the resulting map is generated and displayed for the user.

[0077] In yet another embodiment, the user may request a map that includes both personal and general placemarks 430. The client receives 432B a map and a group of general placemarks or entities associated with the geography of the map based on this request. The general placemarks are combined with personal placemarks that have been designated by the user in any of a variety of ways (for instance by authoring the entity definition or placing the placemark in a favorites folder), and a ranking mechanism is applied 436B to the combined group of placemarks. The results are used to generate 440B a layer that includes both personal and generalized placemarks, which is then combined with the map provided to the client to generate 450B a map for display 450B.

[0078] FIG. 5 shows an exemplary map including personal and general placemarks that has been generated in accordance with the method described at steps 412-420 and 430B through 450B. As is shown in FIG. 5, personal placemarks including Max’s playground 105 and Dirk’s coffee spot 419 are depicted together with dining (e.g. 510A) and drinking placemarks (e.g. 510B) on the map 550. Also provided are community placemarks which are demarcated with a special community placemark icon 540. The map 550 also includes interactive content, in the form of a text box 530 that links to a comment about a placemark provided by a community member. This allows a user to benefit from the input provided by other community members.

[0079] The features and advantages described herein are not all-inclusive and, in particular, many additional features and advantages will be apparent to one of ordinary skill in the art in view of the figures and description. Moreover, it should be noted that the language used in the specification has been principally selected for readability and instructional purposes, and not to limit the scope of the inventive subject matter.

PPC Advertising Campaign Strategy

 1)      Allows us to prepare campaigns and fully leverage our time effectively

     a.       Ad design

     b.      Keyword research

     c.       Keyword bid levels

2)      Choosing the right media channels

3)      Turning on Campaign Optimization and Setting Monthly Budgets

4)      Goaling a Campaign (determine # of ads to deliver and amount of time in which you want to deliver them)

5)      Adjusting Campaign Monthly Budgets

6)      Setting Campaign Daily Spend Limits

7)      Understanding Campaign Schedules

      Coordinating team effort campaign deployment on channels

Free and Affordable Analytics Solutions

• Woopra
• Blvd Status
• Clicky
• Stat24
• Etracker

Search Analytics

• HitTail
• 103bees
• Enquisite
• Raven
• Compete

Heat Maps and Usability Tracking

• Crazy Egg
• Clicktale
• clickdensity
• Clicktracks
• Fuse Stats

Enterprise Analytics

• Omniture
• Web Trends
• IndexTools/Yahoo! Web Analytics
• CoreMetrics
• Unica

Hosted/Server Side Solutions

• Piwik
• Mint
• Trace Watch
• AW Stats
• Analog

Twitter Analytics (for Short URLs)

• HootSuite
• bit.ly
• Cligs
• Poprl.com
• Zi.ma

Social Media Analytics

• Radian6
• onalytica
• NetEquity
• BuzzLogic
• FirstRain
• Ethority
• Brandwatch
• converseon
• CustomScoop
• Attensity

we recently ran some tests dissecting some campaigns and creating 8 Google AdWords campaigns from one. We wanted to see whether traffic would be decreased or increased.  Our test confirm breaking down campaigns and getting as granular as you can definitely pays off. It’s better to organize multiple keywords under various campaign groups.

Use the [Exact Match] if possible

Google AdWords lets you set the following phrase-matching options when you are entering the keywords you want to bid on:

• Broad match: This is the default option, and I’d say, the most inefficient way of selecting the search options. Here, if you enter content management software it will show your ad for all possible combinations of the three words, plus, if they are used in conjunction with some other word. So whether it is management software or content software or buy content or something else, your ad will appear for all of them, attracting tons of unwanted clicks and hence, expense.
• Phrase match: In this phrase match option your ad appears for the phrase match plus for any other word appearing with the phrase. So if you use “content management software” (with double quotes) it will not only show your ad for content management software but also for free content management software, download content management software, download content management software, content management software in php, etc.
• Exact match: This is the best way of entering your keywords and it can get you the highest conversion rate (although fewer links). For exact match you enclose your keywords within [ and ]. Hence, if you enter [buy content management software] your ad will only show up for this search term and nothing else. This drastically reduces the clicks you get but since people come to your website through the exact search term the conversion rate is quite high.
• Negative match: Suppose you don’t want to appear for the word free If you use -free content management software your ad won’t appear for the search terms containing the string free. Use this option to mark for all the keywords and phrases you don’t want to appear for.

We analyzed and checked the numbers for the same keywords in generic (r7168) and specific (r9100) ad groups. Below is what we have found. Specific ad groups actually get a bit more impressions.

Month	Keyword		Data	r7168	r9100
January	“diapers coupon”		Sum of Impressions	133	5
			Sum of Clicks	0	0
			Sum of Conversions	0	0
	“diapers coupons”		Sum of Impressions	508	335
			Sum of Clicks	3	2
			Sum of Conversions	0	0
	[diapers coupon]		Sum of Impressions	85	5
			Sum of Clicks	0	0
			Sum of Conversions	0	0
	[diapers coupons]		Sum of Impressions	21	159
			Sum of Clicks	0	0
			Sum of Conversions	0	0
	diapers coupon		Sum of Impressions	134	100
			Sum of Clicks	1	0
			Sum of Conversions	0	0
	diapers coupons		Sum of Impressions	10	318
			Sum of Clicks	0	2
			Sum of Conversions	0	1
January Sum of Impressions				891	922
January Sum of Clicks				4	4
January Sum of Conversions				0	1
December	“diapers coupon”		Sum of Impressions	42	7
			Sum of Clicks	0	0
			Sum of Conversions	0	0
	“diapers coupons”		Sum of Impressions	171	893
			Sum of Clicks	2	7
			Sum of Conversions	1	0
	[diapers coupon]		Sum of Impressions	18	3
			Sum of Clicks	0	0
			Sum of Conversions	0	0
	[diapers coupons]		Sum of Impressions	16	65
			Sum of Clicks	0	1
			Sum of Conversions	0	1
	diapers coupon		Sum of Impressions	85	180
			Sum of Clicks	0	0
			Sum of Conversions	0	0
	diapers coupons		Sum of Impressions	20	198
			Sum of Clicks	1	3
			Sum of Conversions	1	0
December Sum of Impressions				352	1346
December Sum of Clicks				3	11
December Sum of Conversions				2	1

Given the economic downturn, digital marketing tactics are shifting and businesses are being forced to take a hard look at which channels should be a part of their 2009 marketing plan, and to what extent.

According to a study put out by Marketing Sherpa last month, only Web 2.0 (social network marketing) and email marketing are expected to become a larger part of digital marketing plans moving into ’09, while paid search marketing (PPC), display advertising, mobile marketing, and other forms of digital advertising will be phased out to a certain extent.

Marketing Sherpa goes on to explain that various channels may still experience overall growth as the study accounts for those that will receive more or less attention, but NOT the budget which will be allocated toward or away from that channel.

For example, about 1/3 of companies are expected to reduce PPC spending; however, it is estimated that their reduction in spend will be trumped by the amount of increased spend of the 27% of companies that plan to ramp up their PPC efforts in ’09.

Regardless, though, of whether your marketing budget is increasing, being slashed or remaining static, given the times, every dollar spent in the coming year will likely come under much more scrutiny. While tactics may be shifted across the board, I want to talk briefly about paid search marketing given the inherent flexibility of this channel.

If you’re a part of one of the 33% of companies planning to cut your PPC budget, you or your agency may be in the midst of some serious analysis. What’s working? What’s not working? How do we best shift budget?

Many companies will no longer have the luxury of prospecting as much in this channel. Strategies will need to be more targeted to reduce wasted spend. If brand awareness was previously at the forefront, it may take a back seat to ROI moving forward. And while trial and error will still represent the foundation for campaign performance and efficiency, there may be fewer trials and less room for error.

More than ever, it’s going to be important to take a close look at every keyword and every ad, figure out what deserves the most attention, and how best to reach your audience.

For some, long tail will become a bigger part of their overall keyword strategy. For others, top-level keywords will still be an important part of their branding efforts. Rather than reaching your audience in all phases of their shopping experience, maybe you can only afford to try and capture those that have done the research and are ready to purchase.

The extent to which strategies need to be refined will vary from company to company; however, a commonality between all PPC marketers will be the need to adjust performance metrics on the fly. Despite whether or not your budget is changing, the factors driving the purchasing decisions of your target market likely are changing, and they need to be taken into account when formulating strategy and setting expectations.

Paid search represents an ideal marketing channel to exploit in order to prove much-needed value in today’s economy. The challenge moving forward, however, is figuring out how to get more with less.

I recently started using twitter and realized that setting up accounts with keywords was a brilliant way to lock out my competition in certain verticals. The next part of the equation was how to monetize the twitter account.

* Direct Advertisement – If a lot of people are following you in ‘twitter’ or in other words if you have a ‘wide reach’ to a large number of twitter users, you can look into direct advertisements for monetization.

* Recommendation – You can recommend a particular product or service on behalf of your advertisers & peruse your followers to take an ‘action’.

* Promotion – You can help your advertiser to promote a new product or service to your twitter followers. I have found a huge increase in my Google analytics where twitter is the referring site.

Twitter works like niche marketing, if you are an Internet Marketer, most of your followers in Twitter are probably from the same sphere of influence which is why they have ‘followed’ you in the first place.

# Affiliate Marketing – I was listening to Shoemoney’s podcast from webmasterradio.fm where he talked about how affiliate marketing can actually be blended within Twitter.

* Affiliate Links – If you are recommending a product or a service to your followers, you can always embed your affiliate link in your tweet. (e.g. Using a service like TinyUrl.com)

* Sales Pages – Instead of taking your followers to another website using your affiliate link, you might just want to promote your own product & services and sell them directly via your sales / landing page.

# Paid Review– This is pretty much like the direct advertisement, however the only difference would be, instead of blindly suggesting a product or a service to someone, you can actually do some ‘paid reviews’. Similar to Blogvertise and other paid inclusion sites.

* Product – Paid review about a particular product (e.g. Internet Marketing software)
* Services – Paid review on services that your advertiser offers. (E.g. Mortgage, Auto, pick any affiliate niche)
* People – Paid review about people. You can review someone’s web design skills, or you may promote someone’s ‘twitter’ account for instance.

# Sponsored Contest– An advertiser may want to run a contest via a popular twitter user to get feedback about his product & services or just to create some brand awareness. (gearing up to the buzz marketing campaign)

* Feedback – Example – “What New features would you like to see in Product A”?
* Brand Awareness– Example- “Describe product X in 140 Characters” or “Tell us why you like product X within 140 Characters”. The best answer can be given a cool prize sponsored by the advertiser.

Search Engine Optimization (SEO) is a critical component in having a successful website! No matter how great your website looks, if it can’t be found on search engines like Google and Yahoo, then its value weakens. Optimization allows your site to be found through specific keyword phrases (META tags) and can raise your website’s ranking to higher positions on top search engines (Google, Yahoo, MSN, etc.) and will greatly increase web traffic exposed to your site. All SEO packages, except the Basic SEO, include monthly maintenance. This is important because people’s search phrase patterns change, the search engine firms’ ranking algorithms evolve, plus all of the major search engines take 30 to 90 days to “crawl/spider” your site to check for updates and optimization changes and if your site hasn’t changed between crawls your ranking will fall; therefore it takes some time to see a significant rise in your website’s ranking. We can also work to build inbound links from other relevant but non-competing sites. The search engines are continually changing their search algorithms and changing the rules on what constitutes ‘spamming’. Many cheaper SEO firms use automated submission and optimization techniques, while faster, the major search engines are aware of this and will penalize your site (with low rankings) if these unethical methods are used. We only use Organic SEO techniques, otherwise known as Natural search engine optimization, that guarantee you won’t get penalized while achieving the high search engine ranking your business deserves.Other terms for Search Engine Optimization are:

* SEO
* Organic Search Engine Optimization – Organic SEO
* Natural Search Engine Optimization – Natural SEO
* Search Engine Positioning
* Search Engine Marketing – SEM
* Search Engine Ranking Optimization
* Internet Marketing

Business Basic SEO Package – $355/month
(12 month minimum) The Business Basic search engine optimization package is designed for small businesses who want to develop their presence on the internet and increase the amount of traffic driven to their website. To make your website optimization successful, you
must be ever-mindful of your competition and their marketing efforts. In order to be found on search engines, the need for a well optimized website is paramount. To help you keep track of your competition, we have included a Competition Analysis Report with this package. Giving you a baseline to see how you stand-up against your competition. This package includes ongoing analysis, research, optimization, submissions as well as delivery of a Competition Analysis Report and detailed Search Engine Ranking Reports, to keep your website in a high ranking spot on major search engines.INCLUDES:

* Optimized for up to 10 keyword phrases on up to
20 pages.
* Top 5 Competition Analysis Report
We will send you a complete review of your website as well as a review of your top 5 competitors’ websites. Giving you a baseline to see how you stand-up against your competition.
* One Initial Search Engine Ranking Report, plus
2 bi-annual Reports.
* Analysis of Keyword popularity
* Analysis of link popularity, internal link text & inbound links
* Optimization of web page titles
* Optimization of Meta Keywords & Descriptions
* Optimization of the body text & headlines
* Optimization of Img Alt attributes for all images
* Optimization of the internal link structure
* Hand Submissions to major engines (including Google, Yahoo, MSN)
* Hand Submissions to major directories
* Submission to 40 minor search engines
* Submission to 20 minor directories & special interest directories
* Re-Submissions to keep you at the top
* Re-optimization and tweaking to enhance positions

Business Plus SEO Package – $475/month
(12 month minimum) The Business Plus search engine optimization package is suitable for any website owner or company who wants to significantly develop their presence on the internet, raise their search engine ranking and increase their market share. This package is designed for small to medium sized businesses, and is the smallest plan recommended for e-commerce websites, who would like to increase website traffic, sales and awareness of their brand to a broader audience.
Since this market is larger, the number of competitors you have increases. Therefore, in order to be found on search engines, the need for a well optimized website is paramount. Top 10 Competition Analysis Reports and Quarterly delivery of detailed Search Engine Ranking Reports are included, to continually keep you informed and in a high ranking spot.
With the ever growing number of websites, changes in search engine algorithms and your competitors marketing efforts, you have to continuously update and optimize your site to maintain a high ranking in the top search engines. INCLUDES:
(Items in White text are unique to this package)

* Optimized for up to 20 keyword phrases on up to
40 pages.
* One Initial Report, plus 4 Quarterly Search Engine Ranking Reports
* Top 10 Competition Analysis Report – We will send you a complete review of your website as well as a review of your top 10 competitors’ websites.
* All body copy will be rewritten by an SEO specialist to assure high search engine ranking for key phrases.
* Submission to 60 minor search engines
* Submission to 40 minor & special interest directories
* Creation of a spider friendly site map
* Reciprocal Linking to Increase Link Popularity
* Deep Submission strategy to make sure even the inner pages are indexed
* Hand Submissions to major engines (including Google, Yahoo, MSN)
* Hand Submissions to major directories
* Analysis of Keyword popularity
* Analysis of link popularity, internal link text & inbound links
* Optimization of web page titles
* Optimization of Meta Keywords & Descriptions
* Optimization of the body text & headlines
* Optimization of Img Alt attributes for all images
* Optimization of the internal link structure
* Re-Submissions to keep you at the top
* Re-optimization and tweaking to enhance positions

Business Pro SEO Package – $625/month
(12 month minimum) The Business Pro search engine optimization package is designed for any company who wishes to create and sustain
a strong long term presence on the internet. This package covers every aspect of the search engine optimization campaign essential to the success of your business.
It is designed to work closely with your marketing plan to increase your revenue and market share. This is also the package we suggest for e-Commerce sites in order to drive traffic to their site and increase the bottom line.
This package includes everything in the Business Plus package, PLUS a continuous link building campaign, quarterly Competition Analysis reports, monthly Search Engine Ranking reports to continually keep your site in a high ranking spot. INCLUDES:
(Items in White text are unique to this package)

* Optimized for 35 or more keyword phrases on up to 50 pages.
* One Initial Report, plus 12 Monthly Search Engine Ranking Reports
* Bi-annual Top 10 Competition Analysis Reports -
We will send you a complete review of your website as well as a review of your top 10 competitors’ websites every 3 months. Giving you a expanded view to see how you stand-up against your competition over time.
* Active Link Building Campaign to Increase Link Quality and Popularity
* All body copy & product descriptions will be rewritten by an SEO specialist to assure high page ranking.
* Submission to 75+ minor search engines
* Submission to 60 minor & special interest directories
* Creation of an XML format spider friendly site map
* Deep Submission strategy to make sure even the inner pages are indexed
* Hand Submissions to major engines (including Google, Yahoo, MSN)
* Hand Submissions to major directories
* Analysis of Keyword popularity
* Analysis of link popularity, internal link text & inbound links
* Optimization of web page titles
* Optimization of Meta Keywords & Descriptions
* Optimization of the body text & headlines
* Optimization of Img Alt attributes for all images
* Optimization of the internal link structure
* Re-Submissions to keep you at the top
* Re-optimization and tweaking to enhance positions

Basic SEO Package – $985 one time service The Basic search engine optimization package takes the guess work out of search engine optimization and is suitable for a website owner familiar with SEO techniques who can perform SEO maintenance on their own, therefore does not include monthly maintenance. INCLUDES:

* Optimized for up to 10 keyword phrases on up to 10 pages
* Initial Search Engine Ranking Report (additional reports available)
* Analysis of Keyword popularity
* Analysis of link popularity, internal link text & inbound links
* Optimization of web page titles
* Optimization of Meta Keywords & Descriptions
* Optimization of the body text & headlines of pages
* Optimization of Img Alt attributes for main images
* Hand Submissions to major engines (including Google, Yahoo, MSN)
* Hand Submissions to major directories
* Submission to 10 minor search engines

fabbingmachines.com Such machines could evolve from the 3-D printers currently used by industrial engineers for “rapid prototyping.” They design parts in computer-aided design programs and feed the designs to 3-D printers to make working plastic models. A 3-D printer has a small nozzle that scans back and forth across a surface, depositing tiny droplets of quick-hardening plastic. After each scan, the nozzle moves up a notch and scans again until it has built up the complete object, layer by layer. With multiple nozzles or a means of swapping supply cartridges, the machine can create objects made of many different materials. An electronic circuit, for example, can be made by combining an organic semiconductor, metallic inks and ceramic insulators.