CN105917335B - Discover viewsheds and vantage points by mining geotagged data - Google Patents

Abstract

An architecture that obtains and utilizes a collection of geotagged data to find an optimal vantage point for a field of view of an entity of interest, such as a physical entity, and a conceptual entity, such as a landmark, sunset, skyline, and the like. The disclosed architecture discloses utilizing image data tagged with at least geotags to discover relationships between combinations of concrete entities and/or abstract concepts, and techniques for exposing those relationships to users. The data may be crowd-sourced geo-tagged image data that may be mined from social content and viewed or experienced from a certain location/area.

Description

Discover viewsheds and vantage points by mining geotagged data

Background technique

Most major cities have iconic landmarks on the "to-do" list of tourists visiting the city. Lists of these landmarks and associated locations are readily available online or in print. What is not available, however, is a vantage point looking at these landmarks (where the user can take a good photo of the landmark), or in general, the relationship between this landmark and other locations/areas and how such relationships are valid conditions of.

This coveted information is often limited by the local community's knowledge and cannot be easily discovered through traditional sources available to people unfamiliar with the city. Also, this is the type of information that people don't know they'll want or need before visiting, but would be worth knowing before visiting.

Contents of the invention

The following presents a simplified summary in order to provide a basic understanding of some of the novel embodiments described herein. This summary is not an extensive overview and it is not intended to identify key/critical elements or delineate their scope. Its sole purpose is to present some concepts in a simplified form as a prelude to the more detailed description that is presented later.

The disclosed architecture obtains and utilizes aggregated (e.g., crowdsourced) geotagged data (files tagged with geographic location information, or "geotagged" files, such as images) to discover objects targeting entities of interest (e.g., The best vantage point for the viewshed of physical entities and conceptual entities). Physical entities may be landmarks and other points of interest, and abstract concepts may be sunsets, seashores, skylines, street art, storms, meteor showers, etc., which do not necessarily have an associated physical location.

The disclosed architecture discloses utilizing at least geo-tagged image data to discover relationships between concrete entities and/or combinations of abstract concepts, and techniques for exposing these relationships to a user. More specifically, this involves utilizing crowdsourced geotagged image data to discover vantage point views for both existing entities and abstract concepts (e.g., sunsets, "romantic" landscapes, etc.) Vantage points are mined from social content and can be viewed or experienced from a certain location/area.

The architecture also detects the conditions under which these relationships are active (will positively, negatively affect the user experience, or not affect at all). The vantage point relationship can also be based on external factors such as time of day, day of week, season of year (e.g., summer versus winter), weather conditions, traffic conditions, construction conditions, etc. of. For example, a given vantage point may no longer be feasible when there is fog or the vantage point is so far away from the field of view that inhibits a good view. Similarly, some venues may not be lit enough or not at all at night. Thus, a favorable viewpoint relationship between two locations may only be maintained based on certain conditions (eg, only for certain hours of the day, etc.).

The architecture can also leverage data based on the amount of geotagged data that mentions the vantage point, and the credibility of the user generating the data (e.g., geotagged data from professional photographers is more Higher confidence) popularity scores to enhance vantage point information (or in general, any relationship).

The architecture can also leverage sentiment scores derived from text with geotagged data to enhance relationships (e.g., receiving a response like "Amazing view of the Space Needle" compared to "Amazing view of the Space Needle") The vantage point score of the review of the class is higher).

The framework can also build a graph of mixed entities (physical concepts and abstract concepts) and visualize the structure to the user so that the user can browse the structure to discover new entities (eg, places to visit, as well as abstract concepts).

The architecture facilitates the serendipitous discovery of entities to watch/tour/watch by pushing recommendations to users. The recommendation may be based on the mined relationships and the honor mined conditions for those relationships. For example, when a user is on a ferry going to a particular destination, the architecture can recommend to the user when to turn one hundred and twenty degrees to the left at a particular location for a view of a stunning mountain. The condition could be to make a recommendation only if the sky is clear (condition based on weather) and if the user is a tourist/visitor (condition for user interest grouping).

When exploring/searching for a particular entity, the architecture also generates recommendations of related entities (eg, concepts) to the user. For example, when a user is using a search engine to search for a particular trip (e.g., a cruise), and while the user is still at home, other entities related to that trip (e.g., great views of mountains and beaches) can be generated and presented possibility) recommendations.

To the accomplishment of the foregoing and related ends, certain illustrative aspects are described herein in conjunction with the following specification and drawings. These aspects are indicative of various ways in which the principles disclosed herein may be implemented and all aspects and equivalents thereof are intended to be within the scope of the claimed subject matter. Other advantages and novel features will become apparent from the following detailed description from the following detailed description when considered in conjunction with the accompanying drawings.

Description of drawings

FIG. 1 illustrates a system for discovering horizons and vantage points according to the disclosed architecture.

FIG. 2 illustrates an alternative system for discovering horizons and vantage points according to the disclosed architecture.

Figure 3 shows a flowchart for generating viewshed clusters.

4 shows a map for mining viewsheds and vantage points for landmark entities such as the Space Needle using geotagged photo data.

FIG. 5 shows a map for mining viewsheds and vantage points for conceptual entities based on geo-tagged "birds" data.

Figure 6 shows a graph that can be constructed based on the generated word frequency histogram.

Figure 7 illustrates a method according to the disclosed architecture.

Figure 8 illustrates an alternative approach in accordance with the disclosed architecture.

9 shows a block diagram of a computing system that performs discovery of viewsheds and vantage points by mining geo-tagged data in accordance with the disclosed architecture.

Implementation

Knowledge of the best view (e.g., from a photographic point of view) or vantage point for viewing objects and points of interest (known as the viewshed) is desirable (especially for those who do not frequently visit the location user). Such information has a high user pleasure factor when presented at the right moment, and it can be used in any new geographic scenario that intends to delight the user by suggesting things to do and places to visit in the city .

For example, the Space Needle is an iconic Seattle landmark. However, standing next to the Space Needle will not produce the iconic imagery usually associated with this landmark. Revealing where locals took pictures of the Space Needle in Seattle can greatly enhance the experience for newcomers who want to capture cherished images and memories. Another example is discovering vantage points for more abstract concepts that the user would be interested in, eg, good places to watch sunsets, bird watching, street art, etc. At the same time, the conditions that hold the relationships between a set of entities/concepts need to be mined and made available to users. For example, the relationship "location A is a good vantage point for location B" may only be valid during certain times of day (e.g., only during daytime), or only under certain weather conditions (e.g., clear skies are required to See Rainer Mountain) is available.

The disclosed architecture acquires and utilizes crowdsourced geotagged data (files tagged with geographic location information, or "geotagged" files, such as images) to discover the best view-sheds for physical test questions and conceptual entities. A good vantage point, where physical entities may be landmarks and other points of interest, and abstract concepts such as sunsets, seashores, skylines, street art, storms, meteor showers, etc., do not necessarily have an associated physical location.

By applying word similarity measures, concept hierarchies/categories, and user locale data, aliases (and misspellings) can be detected, descriptions can be combined or generalized, and labels and names can be localized, enabling use of relevant view (popular sunset viewpoints, beach viewpoints, etc.) in response to user actions (eg, a search for "romantic restaurants").

The disclosed architecture discloses utilizing at least crowdsourced geotagged image data to discover relationships between concrete entities and/or combinations of abstract concepts, and methods for exposing such relationships to users. More specifically, this includes leveraging crowdsourced geotagged image data to discover viewsheds (for existing entities as well as mined from social content and that can be viewed or experienced from a location/area). A vantage point for an abstract concept (eg, both a sunset, a "romantic" landscape, etc.). For example, Alki Beach in Seattle is a vantage point for the Space Needle, and Alki Beach is an experience point for the concepts of "romantic" and "fresh sea breeze".

In other words, the disclosed architecture employs spatial and/or spatiotemporal clustering of user-defined geotagged image content, identification of distinct entities, and association of common and related entities presented to users, representations of entities of interest based on what people have said, using concept-graphs to classify geotagged image content (including temporal classification and many-to-many relationships), and A serendipitous discovery of something to visit. User profile information and/or inferences (eg, tourist vs. local resident) may also be incorporated.

Spatial and/or spatiotemporal clustering of user-defined labels can include concrete entities (e.g., "Space Needle") but also more diffuse/abstract concepts ("sunset", "art", etc.) . They can be organized in concept-graphs (eg, "romantic", "autumn", "quirky", etc.) using many-to-many relationships (eg, "sunset" is "romantic" and is "natural beauty"). The architecture assists the user in the map exploration experience, such as helping the user discover local features of an area or a serendipitous discovery of something to see or visit.

The architecture also detects the conditions under which these relationships are valid (will positively, negatively affect the user experience, or not affect at all). The vantage point relationship may also be based on external factors such as time of day, day of week, season of year (eg, summer versus winter), weather conditions, and the like. For example, Alki Beach is no longer a vantage point for the Space Needle when there is fog or the vantage point is too far away from the Space Needle to inhibit a good view. Similarly, some venues may not be lit enough or not at all at night. Thus, a favorable viewpoint relationship between two locations may only be maintained based on certain conditions (eg, only for certain hours of the day, etc.). The architecture can also leverage time-tagged data alone or in combination with geo-tagged data.

The architecture can also leverage data based on the amount of geotagged data mentioning that vantage point and the trustworthiness of the user generating that data (e.g., professional photographers have more geotagged data than non-professionals). High confidence) popularity scores to enhance vantage point information (or any relationship in general). Credibility can also be based on whether the "author" (the creator of the content (e.g., a photo)) is in the user's social network (e.g., "is a friend", "liked", or is commented on by one user to another user). other well-known social networking tags applied). Additionally, the rating/presentation may be based on whether a friend of the user took the photo or has been to the location depicted in the photo.

The architecture can also leverage sentiment scores derived from text with geotagged data to enhance relationships (e.g., receiving a response like "Amazing view of the Space Needle" compared to "Amazing view of the Space Needle") The vantage point score of the review of the class is higher). Other examples of sentiment scores include, but are not limited to, romantic, gloomy, and more complex attributes other than just "good" and "amazing." Additionally, other scores may relate to likes/dislikes and star ratings where the platform supports these attributes.

The framework can also construct hybrid entity and abstract concept-graphs and visualize the structure to the user so that the user can browse the structure to discover new entities (eg, places to visit, as well as abstract concepts).

The architecture facilitates the serendipitous discovery of entities to watch/tour/watch by pushing recommendations to users. The recommendation may be based on the mined relationships and the conditions of reputation mining for such relationships. For example, when a user is on a ferry bound for a particular destination, the architecture can recommend to the user when to look at a 120-degree angle to the left at a particular location (e.g., with a favorable viewpoint relationship to Mount Rainer). Breathtaking views of Rainer Mountain. The condition could be to make a recommendation only if the sky is clear (condition based on weather) and if the user is a tourist/visitor (condition for user interest group).

When exploring/searching for a specific entity, the architecture also generates recommendations about entities (eg, concepts) to the user. For example, when a user is using a search engine to search for a particular trip (e.g., a cruise to the village of Tillicum), and while the user is still at home, information about other entities related to that trip (e.g., seeing Mount Rainer and Alki Possibility of very good view of beach) is recommended.

The architecture applies machine learning techniques (e.g., clustering) to identify geotagged data (e.g., , image) and denoise the data.

Typically, whitelists (accepted or approved words) are described without specifying the origin of landmarks and concepts (e.g., curated from search logs, mined, derived from unsupervised learning, etc.) steps without loss of generality. A given entity (eg, a landmark or a concept) can be known by multiple aliases (combined or "joined" before clustering) or in a concept hierarchy (joined after clustering). Since the processing of world-wide imagery is expected to use distributed parallelism, it is recommended to stitch clusters (of points) across grid seams (map gridlines). Thus, programming models employing map and reduce functions (e.g., MapReduce) that enable distributed and parallel processing of map and reduce functions (e.g., localized computer nodes to computing systems can be employed and/or clustering of geographically distributed grids).

This can be done by examining the text data associated with a given crowdsourced item (e.g., a tag, a comment, a message such as a "tweet", etc.), using the reverse geocoder API (Application Programmable Interface), or their to reverse-geocode (convert point locations defined by latitude/longitude coordinates into readable addresses or place names) the resulting clusters of data points. Once the clusters have been reverse-geocoded, the resulting locations are assigned as vantage points for entities of interest (eg, original concepts or landmarks). When extracting clusters based on existing locations (eg, landmarks), one of the clusters will most likely be the actual landmark itself. The disclosed architecture can be applied to any crowdsourced data including geographic coordinates and geotagged data such as images.

As an alternative to reverse geocoding, a location may be identified by a more general description such as a vantage point of sunset that is just a bend in a particular road and not represented in the entity set used for reverse geocoding. The latitude/longitude (latitude and longitude coordinates) of the location are considered good photographic vantage points.

In an alternative implementation, the architecture may operate to simply ask users to share their best view, without analyzing and detecting geotagged data.

Referring now to the drawings, wherein like reference numerals are used to refer to like elements throughout. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the present invention. It is evident, however, that the novel embodiments may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form in order to facilitate their descriptions. The present invention is to cover all modifications, equivalents, and alternatives falling within the spirit and scope of the claimed subject matter.

FIG. 1 illustrates a system 100 for discovering horizons and vantage points according to the disclosed architecture. System 100 can include analysis component 102 configured to analyze geotagged data 104 associated with entity of interest 106 and discover relationship 108 between location-based entity 110 and entity of interest 106 . Geotagged data may be crowdsourced image data, eg, photos accessible via social networks. An entity of interest 106 may be an object (eg, a landmark, building, etc.) or an abstract concept (eg, a sunset). The system 100 also includes a viewing generation component 112 configured to generate, based on the discovered relationships 108, vantage point information 114 from which to view the entity of interest 106.

FIG. 2 illustrates an alternative system 200 for discovering horizons and vantage points in accordance with the disclosed architecture. The system 200 may include one or more of the components of the system 100 of FIG.

The augmentation component 202 can be configured to augment the vantage point information with popularity data and source credibility data. Alone or in combination therewith, the augmentation component 202 can be configured to augment vantage point information with sentiment data derived from association with geotagged data.

The recommendation component 204 can be configured to recommend new entities of interest based on relationships and relationship conditions as part of exploring entities of interest. Condition component 206 can be configured to detect spatial and/or temporal conditions for which the relationship is valid, eg, time of day and location at that time. Mapping component 208 can be configured to generate a graph (eg, histogram, node links, etc.) of entities of interest and relationships between entities of interest to discover new entities. Geotagged data can be obtained from social networks, and analysis component 102 can employ machine learning techniques to identify and denoise the geotagged data.

Although not shown, systems 100 and 200 may also include a privacy component that enables authorized and secure processing of user information (eg, geotagged image data). The privacy component enables users to opt-in or opt-out of tracking information as well as personal information that may have been obtained upon registration and utilized thereafter. The privacy component also ensures proper collection, storage, and access to user information, while allowing selection and presentation of content, features, and/or services that facilitate the benefit of a richer user experience and access to more relevant information.

It should be understood that certain components may be rearranged, combined, omitted, and additional components may be included in the disclosed architectures. Additionally, in some embodiments, all or some of these components may be presented on the client, while in other embodiments some components may reside on the server or be provided by local or remote services.

FIG. 3 shows a flowchart 300 for generating viewshed clusters. First, a grid-based term frequency histogram (TFH) 302 is received at 304 for processing. Flow goes to 306 to perform word processing for semantic similarity using, for example, pairwise word processing (e.g., Levenshtein distance) to determine whether a word in TFH is a common misspelling word, abbreviation, obscenity, etc. . At 308, a blacklist comparison operation is performed on the set of words output from the word processing at 306 to remove blacklisted words 310 (unacceptable or rejected words). Black-named words 310 are updated as part of the processing in graph 300 and fed back for use with the new set of words in TFH.

Once the blacklist processing is complete, flow goes to 312 to process 314 the whitename words. White-named words 314 are those words specifically searched for or expected to be in the TFH, and the confidence in these words is increased. At 316, alias list processing is performed to find words that may be shortened versions, slang versions, etc. At 318, clustering is performed on the current state of the word list to find one or more clusters of data points. At 320, a cluster join operation is performed in an attempt to determine whether any of the clusters from previous steps can be joined or are a different set.

Once the clusters are determined, flow goes to 322 to check if one or more clusters are entity locations. If so, flow goes to 324 to identify the cluster as an entity cluster, if not, flow goes to 326 to identify the cluster as a viewshed cluster. In either case, flow goes to 328 to then perform stitching across the grid by various types of grid-based machines. At 330, confidence processing is performed. If the confidence is low (below a threshold), it may be the case that the cluster can be ignored or wait until more data points have been obtained and the confidence is increased before being published. Optionally, flow can then go to 310 to include words on the blacklist. If the confidence is high (equal to and above the threshold), the cluster points are sent 332 for a back-encoding process, identifying either for each general cluster, or for each data point in a cluster one or more addresses. At 334, the viewshed clusters are output.

Below is an example where each point on the map represents a separate geotagged photo. Points depicted similarly (eg, with the same color) fall into the same cluster Cx, while outlier points Oy are identified as being outside the cluster.

FIG. 4 shows a map 400 for mining viewsheds and vantage points for landmark entities such as the Space Needle using geotagged photo data. Each point on map 400 represents a single instance of photo data. The architecture has identified seven vantage points from which to view the Space Needle (SN): the zeroth vantage point associated with the zeroth cluster C0, the first vantage point associated with the first cluster C1, the second The second vantage point associated with cluster C2, the third vantage point associated with the third cluster C3, the fourth vantage point associated with the fourth cluster C4, the fifth vantage point associated with the fifth cluster C5. A vantage point, and a sixth vantage point associated with the sixth cluster C6. Two points of the plurality of outliers are indicated as O1 and O2.

More specifically, the zeroth vantage point is the Space Needle itself, the zeroth vantage point is defined by a plurality of photos taken close to the Space Tower, and the first vantage point associated with the first cluster C1 may be Parks (e.g., Kerry Park), the second vantage point associated with the second cluster C2 may be another park (e.g., Gas Works Park), the third vantage point associated with the third cluster C3 may be another A park (e.g., Volunteer Park), a fourth vantage point associated with a fourth cluster C4 may be from an observation deck of a building in downtown Seattle, and a fifth vantage point associated with a fifth cluster C5 may be from Beach (eg, Alki Beach), while a sixth vantage point associated with sixth cluster C6 may be from a popular market (eg, Pike's Place Market).

FIG. 5 shows a map 500 for mining viewsheds and vantage points for conceptual entities based on geotagged "bird" data. Each point on map 500 represents a single instance of photo data. The architecture has identified four vantage points from which to view "birds": a seventh vantage point associated with the seventh cluster C7, an eighth vantage point associated with the eighth cluster C8, a ninth vantage point associated with the ninth cluster The ninth vantage point associated with C9, the tenth vantage point associated with the tenth cluster C10. Two of the multiple outliers are indicated as O3 and O4. A cluster may be an "active" cluster where other users have captured images of the entity of interest, and the "active" cluster may be considered a vantage point from which to obtain a desired photo.

More specifically, the seventh vantage point associated with the seventh cluster C7 may be a zoo (e.g., Woodland Park Zoo), and the eighth vantage point associated with the eighth cluster C8 may be a park (e.g., Gas Works park), the ninth vantage point associated with the ninth cluster C9 may be another park (for example, Washington Park Botanical Garden), and the tenth vantage point associated with the tenth cluster C10 may be from a natural ecological area (for example, Union Bay Natural Ecological Area).

By developing a taxonomy based on clustering words, suitable locations can be suggested to people in different interest groups. For example, the vantage point above for the "birds" tag could be offered to bird watchers, or more generally, to someone belonging to the "outdoors" interest category.

Concrete entities, abstract concepts, and associated relationships may also be rendered in the form of a graph that enables a user to discover each of the concrete entities, abstract concepts, and associated relationships. The construction of the graph can be based on the mined histogram. For example, an observation of a word frequency histogram at Denver's Skyline Park may reveal a high popularity of the word "parkour" (an overall training discipline). The fact that this location is a parkour spot is useful in itself, and the map can also be used to show the user other parkour spots in Denver.

FIG. 6 shows a graph 600 that may be constructed based on the generated word frequency histogram. Graph 600 includes nodes and relationships of physical location entities (entities of interest), vantage points as nodes, words for nodes, and links between one or more of the nodes and words. All or portions of graph 600 may be generated and presented to a user via a device display for viewing nodes and relationships. Although depicted as a double arrow for a bidirectional relationship from which discovery can come from either node, it should be understood that some relationships can be unidirectional (discoverable from only one node and not from the node where the relationship exists. found in another node).

Continuing with the "Space Needle" example, graph 600 may include a node 602 for the entity of interest ("Space Needle," or SN for short), a node 604 for Alki Beach ("AB"), a node 604 for Volunteer Park ("VP") Node 606 for Pike's Place Market (“PPM”), Node 608 for Gas Works (“GWP”), Node 612 for Columbia Tower (“CT”), Node 612 for Kerry Park (“KP”) node 614 for , and node 616 for Seattle Center ("SC"). Other nodes and links (relationships) can be provided but not shown.

The graph generation procedure can then be repeated at each new node in the graph (using the location and/or top abstract word) to reveal new related locations and words. In the above example, the expansion of the graph can be triggered at all nodes except the Space Needle (the first originator node of the graph).

According to the above diagram, users can gain insights including but not limited to the following contents. Using social media, people generate information about the Space Needle (node 602) from: Columbia Tower (node 612), Kerry Park (node 614), Alki Beach (node 604), Volunteer Park (node 606), Seattle Center (node 616), Pike's Place Market (node 608), and Gas Works Park (node 610).

The graph 600 indicates that if the user visits the Space Needle, the user should probably check out these other locations as well. During the night ("night"), it might be good to visit Alki Beach or the Space Needle. The user can view the city ("city") and Puget Bay ("puget bay") from both locations.

If the user visits Alki Beach (node 604), the user may meet photographers who take pictures of the beach ("Beach") and cityscapes ("Cityscape"). Alki Beach (Node 604) and Columbia Tower (Node 612) are similar, because people generate views about the cityscape ("Cityscape"), the skyline ("Skyline"), and the downtown area ("Skyline") from these two locations. urban area") data. Therefore, the user may be able to get a good view of the city from these two locations (node 612 and node 604). Alki Beach (node 604) and Space Needle (node 602) may have many edges (links), not all of which are shown, but this indicates that people generate a lot of information about these locations; therefore, users should visit these locations first.

Included herein are a set of flowcharts representative of example methodologies for implementing the novel aspects of the disclosed architecture. Although for simplicity of explanation, one or more methodologies illustrated herein are shown and described as a series of acts, for example, in flow diagram or flowchart form, it is to be understood and appreciated that the methodologies Not limited by the order of acts, as some acts may thereby occur in different orders and/or concurrently with other acts shown and described herein . For example, those skilled in the art will understand and appreciate that a methodology could alternatively be represented as a series of interrelated states or events, such as in a state diagram. Moreover, a novel implementation may not require all acts illustrated in the methodologies.

Figure 7 illustrates a method according to the disclosed architecture. At 700, geotagged data associated with an entity of interest that is an object or abstraction is accessed. At 702, the geotagged data is analyzed to discover relationships between location-based entities and entities of interest. At 704, vantage point information from which to view the entity of interest is generated based on the discovered relationship.

The method may also include deriving conditions under which the relationship is valid. The method may also include augmenting vantage point information with popularity data and source credibility data, and either popularity and/or source credibility. The method may also include utilizing sentiment data associated with the geotagged data to enhance the relationship.

The method may also include constructing and presenting a hybrid graph related to entities of interest to enable discovery of new entities of interest. The method may also include recommending new entities of interest based on relationships and relationship conditions. The method may also include recommending related entities of interest while exploring the entities of interest.

Figure 8 illustrates an alternative approach in accordance with the disclosed architecture. The method may be implemented as a computer-readable storage medium comprising computer-executable instructions, wherein the computer-executable instructions, when executed by a microprocessor, cause the microprocessor to perform the actions in the method. At 800, geotagged data associated with an entity of interest is accessed. The entity of interest may be a physical object or an abstract concept. At 802, the geotagged data is analyzed to discover relationships between location-based entities and entities of interest. At 804, vantage point information from which to view the entity of interest is generated based on the discovered relationship. At 806, hybrid graphs related to entities of interest that enable discovery of new entities of interest are constructed and presented.

The method may also include enhancing vantage point information with popularity data and source credibility data, and enhancing relationships with sentiment data associated with the geotagged data. The method may also include recommending new entities of interest based on relationships and relationship conditions. The method may also include recommending related entities of interest while exploring the entities of interest. The method may also include deriving conditions under which the relationship is valid.

As used in this application, the terms "component" and "system" are intended to refer to a computer-related entity, be it hardware, a combination of software and tangible hardware, software, or software in execution. For example, a component may be, but is not limited to, tangible components such as microprocessors, on-chip memory, mass storage devices (e.g., optical drives, solid-state drives, and/or magnetic storage media drives), and computers, and such as Software components such as procedures, objects, executables, data structures (stored in volatile or nonvolatile storage media), modules, threads of execution, and/or programs that run on a microprocessor.

As an example, both an application running on a server and the server can be components. One or more components can reside within a process and/or thread of execution, and a component can be localized on one computer and/or distributed between two or more computers. The word "exemplary" may be used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as "exemplary" is not necessarily to be construed as preferred or advantageous over other aspects or designs.

Referring now to FIG. 9 , shown is a block diagram of a computing system 900 that performs discovery of viewsheds and vantage points by mining geotagged data in accordance with the disclosed architecture. It should be appreciated, however, that some or all aspects of the disclosed methods and/or systems may be implemented as a system-on-chip, wherein analog, digital, mixed-signal, and other functions are fabricated on a single chip substrate.

In order to provide additional context for various aspects of the present invention, FIG. 9 and the following description are intended to provide a brief, general description of a suitable computing system 900 in which various aspects may be implemented. Although the above description is in the general context of computer-executable instructions that can run on one or more computers, those skilled in the art will recognize that other program modules and/or as hardware and software components can also be combined. combination to achieve novel embodiments.

Computing system 900 for implementing various aspects includes a computer 902 having one or more microprocessing units 904 (also referred to as one or more microprocessors and one or more processors), Computer-readable storage medium (computer-readable storage medium/media also includes magnetic disks, optical disks, solid-state drives, external storage systems, and flash memory drives), and system bus 908 . Microprocessing unit 904 may be any of various commercially available microprocessors, such as single processors, multiprocessors, single-core units and multi-core units of processing circuits and/or storage circuits. Additionally, those skilled in the art will appreciate that other computer system configurations (including minicomputers, mainframes, and personal computers (e.g., desktops, laptops, tablet PCs, etc.), handheld computing devices, microcomputer-based processor or programmable consumer electronic devices, etc., each of which is operatively coupled to one or more associated devices) to implement the novel systems and methods.

Computer 902 may be employed in a data center and/or computing resource (hardware and/or software) supporting cloud computing for portable and/or mobile computing systems (e.g., wireless communication devices, cellular phones, and other devices capable of moving). One of several computers for the computing service. Cloud computing services include, but are not limited to, for example, Infrastructure as a Service, Platform as a Service, Software as a Service, Storage as a Service, Desktop as a Service, Data as a Service, Security as a Service, and API (Application Programming Interface) as a Service.

System memory 906 may include, for example, volatile (VOL) memory 910 (eg, random-access memory (RAM)) and non-volatile memory (NON-VOL) 912 (eg, ROM, EPROM, EEPROM, etc.) computer-readable storage (physical storage) media. A basic input/output system (BIOS) may be stored in non-volatile memory 912 and includes basic routines that facilitate communication of data and signals between components within computer 902, for example, at startup period. Volatile memory 910 may also include high-speed RAM (eg, static RAM) for caching data.

A system bus 908 provides an interface to system components including, but not limited to, system memory 906 to one or more microprocessing units 904 . The system bus 908 can be any of a variety of commercially available bus architectures that can be further interconnected to a memory bus (with or without a memory controller) and a peripheral bus (e.g., PCI, PCIe, AGP, LPC, etc.). Any of several types of bus structures.

The memory 902 also includes a machine-readable storage subsystem 914 and a storage interface 916 for interfacing the storage subsystem 914 to the system bus 908 and other desired computer components and circuits. Storage subsystem 914 (physical storage media) may include one or more of the following storage devices: for example, hard disk drive (HDD), magnetic floppy disk drive (FDD), solid state drive (SSD), flash drive, and/or optical disk storage Drive (for example, CD-ROM drive, DVD drive). Storage interface 916 may include interface technologies such as EIDE, ATA, SATA, and IEEE 1394.

One or more programs and data may be stored in storage subsystem 906, machine-readable and removable storage subsystem 918 (e.g., flash drive form factor technology), and/or storage subsystem 914 (e.g., optical, magnetic solid state), the one or more programs and data include an operating system 920 , one or more application programs 922 , other program modules 924 , and program data 926 .

Operating system 920, one or more application programs 922, other program modules 924, and/or program data 926 may include, for example, items and components in system 100 of FIG. 1 , items and components in system 200 of FIG. 3, the items and processes in diagram 300 of FIG. 3, the map 400 of FIG. 4, the map 500 of FIG. 5, the diagram 600 and relationships of FIG. 6, and the methods represented by the flowcharts of FIGS.

In general, programs include routines, methods, data structures, other software components, etc. that perform particular tasks, functions, or implement particular abstract data types. All or portions of operating system 920, applications 922, modules 924, and/or data 926 may also be cached in memory such as volatile memory 910 and/or non-volatile memory. It should be appreciated that the disclosed architecture can be implemented using various commercially available operating systems or combinations of operating systems (eg, as virtual machines).

Storage subsystem 914 and memory subsystem (906 and 918) serve as computer-readable media for volatile and nonvolatile storage of data, data structures, computer-executable instructions, and the like. Such instructions, when executed by a computer or other machine, may cause the computer or other machine to perform one or more acts of the method. Computer-executable instructions comprise, for example, instructions and data which cause a general purpose computer, special purpose computer, or special purpose microprocessor devices to perform a certain function or group of functions. Computer-executable instructions may be, for example, binary, instructions in an intermediate format such as assembly language, or even source code. Instructions for performing actions may be stored on one medium, or may be stored across multiple media, such that the instructions co-occur on one or more computer-readable storage medium/media, regardless of whether all instructions are on the same medium.

Multiple computer-readable storage media (single media) exclude the propagated signal itself, can be accessed by the computer 902, and include removable and/or non-removable volatile and non-volatile internal and/or external media. For the computer 902, various types of storage media accommodate storage of data in any suitable digital format. Those skilled in the art will appreciate that other types of computer-readable media, such as zip drives, solid-state drives, magnetic tape, flash memory cards, flash drives, magnetic tape cartridges, etc., may be employed for use in implementing the disclosed Computer-executable instructions for novel methods (acts) of the architecture are stored.

A user may interact with the computer 902, programs, and data using external user input devices 928, such as a keyboard and mouse, and through voice commands facilitated by voice recognition. Other external user input devices 928 may include: microphones, IR (infrared) remote controls, joysticks, gamepads, camera recognition systems, stylus, touch screens, gesture systems (e.g., eye movements, e.g. involving hands, fingers, arms , head, etc., body posture), etc. A user may interact with the computer 902 , such as a laptop computer, for example, with onboard user input devices 930 , such as a touchpad, microphone, keyboard, etc., programs, and data using onboard user input devices 930 .

These and other input devices are connected to the microprocessing unit 904 via the system bus 908 through an input/input (I/O) device interface 932, but may also be connected to the microprocessing unit 904 via a parallel port, IEEE 1394 serial port, game port, USB port, etc. , IR interface, short-range wireless (eg, Bluetooth) and other interfaces such as personal area network (PAN) technology to connect. The I/O device interface 932 also facilitates the use of output peripherals 934 such as printers, audio devices, camera devices, etc., such as a sound card and/or onboard audio processing capabilities.

One or more graphics interfaces 936 (also commonly referred to as graphics processing units (GPUs)) communicate between the computer 902 and an external display device 938 (e.g., LCD, plasma) and/or an on-board display device 940 (e.g., for a laptop computer ) to provide image and video signals. One or more graphics interfaces 936 may then be fabricated as part of the computer system board.

Computer 902 may operate in a networked environment (eg, IP-based) using logical connections via wired/wireless communication subsystem 942 to one or more networks and/or other computers. Other computers may include workstations, servers, routers, personal computers, microprocessor-based entertainment appliances, peer-to-peer devices, or other public network nodes, and typically include many or all of the elements described with respect to computer 902 . Logical connections may include wired/wireless connectivity to local area networks (LANs), wide area networks (WANs), hotspots, and the like. LAN and WAN networking environments are commonplace in offices and corporations and facilitate enterprise-wide computer networks, such as intranets, all of which can be connected to a global communications network such as the Internet.

When used in a networked environment, the computer 902 is connected to the network via a wired/wireless communication subsystem 942 (e.g., network interface adapter, onboard transceiver subsystem, etc.) to communicate with a wired/wireless network, wired/wireless printers, wired/wireless The wireless input device 944, etc. communicates. Computer 902 may include a modem or other means for establishing communications over a network. In a networked environment, programs and data associated with the computer 902 may be stored in remote memory/storage devices, such as is associated with a distributed system. It will be appreciated that the network connections shown are exemplary and other elements establishing a communications link between the computers may be used.

The computer 902 is operable to communicate with wired/wireless devices or entities using wireless technologies such as the IEEE 802.xx series of standards, such as being operatively disposed in communication with, for example, printers, scanners, desktop computers and/or portable computers , personal digital assistants (PDAs), communication satellites, any device or location associated with a wirelessly detectable tag (e.g., kiosks, news stations, restrooms), and wireless communications for telephones (e.g., IEEE 802.11 wireless modulation technology) in wireless devices. This includes at least Wi-Fi ^(TM) for hotspots (for validating interoperability of wireless computer networking devices), WiMax, and Bluetooth ^(TM) wireless technologies. Thus, the communication can be a predefined structure like a traditional network or a simple ad-hoc communication between at least two devices. Wi-Fi networks use wireless technology known as IEEE 802.11x (a, b, g, etc.) to provide secure, reliable, and fast wireless connectivity. A Wi-Fi network can be used to connect computers to each other, to the Internet, and to wireless networks (which use technologies and functions related to IEEE-802.3).

What has been described above includes examples of the disclosed architecture. It is, of course, not possible to describe every conceivable combination of components and/or methodologies, but one skilled in the art will appreciate that many further combinations and permutations are possible. Accordingly, the novel architecture is intended to embrace all such alterations, modifications and variations that fall within the spirit and scope of the appended claims. Furthermore, to the extent that the term "comprises" is used in the detailed description or in the claims, such terms are intended to be inclusive in a manner similar to the term "comprises", as when used as a transition word in the claims when "contains" is translated.

Claims (10)

1. A system, comprising:

at least one processor; and

a memory storing instructions that, when executed by the at least one processor, perform the following:

accessing geo-tagged data associated with an entity of interest;

generating a set of location-based entities based on the geotagged data, wherein each location-based entity in the set of location-based entities is a location from which the entity of interest is viewed;

evaluating one or more external factors to determine one or more relationships between at least one location-based entity of the set of location-based entities and the entity of interest to view the entity of interest; and

generating vantage point information from which to view the entity of interest based on the one or more relationships, the vantage point information including one or more location-based entities from the set of location-based entities.

2. The system of claim 1, wherein the memory further comprises instructions for recommending a new entity of interest based on the determined one or more relationships and relationship conditions.

3. The system of claim 1, wherein the memory further comprises instructions for detecting at least one of a spatial condition or a temporal condition that the relationship is valid.

4. The system of claim 1, wherein the memory further comprises instructions for generating a graph of entities of interest and relationships between the entities of interest to discover new entities.

5. The system of claim 1, wherein accessing the geotagged data comprises accessing geotagged data from a social network and using machine learning techniques to identify and denoise the geotagged data.

6. A method comprising the operations of:

accessing geo-tagged data associated with an entity of interest;

evaluating the geotagged data based on one or more external factors to generate one or more relationships between at least one location-based entity of a set of location-based entities and the entity of interest that views the entity of interest, wherein each location-based entity of the set of location-based entities is a location from which the entity of interest is viewed; and

generating vantage point information from which to view the entity of interest based on the discovered relationships, the vantage point information including one or more location-based entities from the set of location-based entities.

7. The method of claim 6, wherein the external factor is at least one of: weather conditions, traffic conditions, and construction conditions.

8. The method of claim 6, further comprising at least one of: augmenting the vantage point information with trending data and source credibility data, or augmenting the relationship with emotion data associated with the geotagged data.

9. The method of claim 6, further comprising: constructing and presenting a hybrid graph related to the entities of interest, the hybrid graph enabling discovery of new entities of interest.

10. The method of claim 6, further comprising: recommending new entities of interest based on the determined one or more relationships and relationship conditions, and recommending related entities of interest while exploring the entities of interest.