Content selection and presentation of electronic content

Abstract

Methods, systems, and apparatus, a method comprising: in response to determining that a threshold number of electronic news article resources each refer to both a first entity and a second entity having a relevance of at least a threshold magnitude: determining that an event likely occurred that includes a particular activity involving both the first entity and the second entity; generating a representation corresponding to the event, generating the representation comprising: generating first content based on one or more of the electronic news article resources that refer to both the first entity and the second entity having the relevance of at least the threshold magnitude, and generating second content based on additional resources, wherein generating the second content based on the additional resources is based on determining that the additional resources relate to an attribute of the particular activity included in the event, wherein the additional resources are resources other than the electronic news article resources that refer to both the first entity and the second entity having the relevance of at least the threshold magnitude, and wherein the attribute of the particular activity is an attribute other than the first entity and the second entity.

Description

[0001] This application is a divisional application of the invention patent application filed on October 25, 2017, with application number 201711008167.2 and title "Content Selection and Presentation of Electronic Content". Technical Field

[0002] This manual covers the selection and presentation of content. Background Technology

[0003] Computing systems can be configured to collect data from multiple web pages and organize the collected data into a searchable index of information. The collected data can be in structured or unstructured formats. For example, a computing system can receive structured or unstructured data for resources including electronic media content, such as articles and other electronic publications. Users consuming media content may want to receive a subset of media content identified as relevant to their specific interests or content preferences. However, many existing systems are not particularly efficient at identifying appropriate content to offer to users. Consequently, existing systems may expend resources identifying, delivering, and displaying content that users will likely ignore. Therefore, there is a need for computing systems that can better utilize computing resources when providing subsets of media content for specific user consumption. Summary of the Invention

[0004] According to the described technique, a computing system receives multiple resources. Resources may include at least a portion of electronic media content and textual content. For each resource, the system may identify one or more entities associated with the resource. Entities may span multiple information types and may include individuals, political entities, entertainment entities, or commercial entities. Entity pairs may be identified by the system, and for each identified pair of entities, the system may determine the number of resources associated with each entity in that pair.

[0005] The system can then determine the occurrence of an event associated with a specific pair of entities. The occurrence of the event can be determined based on the amount of resources associated with each entity in the pair. The system generates a representation corresponding to the event. This representation can be generated based on the resources associated with each entity in the pair. The system can then provide this representation for output to a user device.

[0006] One aspect of the subject matter described in this specification can be embodied in a computer-implemented method. The method includes: receiving a plurality of resources by a computing system, each resource including electronic media content, said electronic media content including at least a portion of text content; for each of the plurality of resources: identifying one or more entities associated with the resource by the computing system; for at least one pair of entities among the identified one or more entities: determining the number of resources in which each entity in the pair is associated.

[0007] The method includes: a computing system determining the occurrence of an event associated with a specific pair of entities based on a determined number of resources; the computing system generating a representation corresponding to the event, the representation being generated based on one or more of the resources associated with each of the entities in the pair; and the computing system providing the representation corresponding to the event to a user device.

[0008] These and other implementations may each optionally include one or more of the following features. For example, in some implementations, the method further includes: detecting specific entity items included in the text content of the resource, the specific entity items being detected based on known user interests; and, in response to detecting the specific entity items, selecting one or more pairs of entities associated with the resource based on the detected specific entity items.

[0009] In some implementations, identifying one or more entities associated with a resource includes: determining at least one interest of a user based on analysis of user data associated with the user, the user data being received during one of the following periods: the current time period or a past time period; and identifying one or more entities associated with the resource based on the determined user's at least one interest.

[0010] In some implementations, determining the occurrence of an event includes: using an event detection module of a computing system to identify multiple candidate events associated with a specific pair of entities; using a filtering algorithm of the event detection module to generate a subset of candidate events associated with resources that include each of the specific pair of entities; and determining the occurrence of the event based on at least one event included in the subset of candidate events.

[0011] In some implementations, identifying one or more entities associated with a resource includes: using at least one collaborative filtering algorithm to determine one or more interests of a first user based at least on the interests of one or more second users; and identifying one or more entities associated with the resource based on the determined interests of the first user.

[0012] In some implementations, identifying one or more entities associated with a resource includes: identifying a first entity based on known user interests; using a relevance detection module of the computing system to determine a second entity associated with the identified first entity; and generating at least one pair of identified entities from one or more entities based on the determined second entity and the identified first entity.

[0013] In some implementations, generating an event-related representation includes: generating a description of the resource, which provides a summary of the electronic media content included in the resource; populating an interest feed to include the resource and the description of the resource, which is used to provide one or more media content items to a user; and using the interest feed, which includes the resource and the description of the resource, to generate an event-related representation.

[0014] In some implementations, the representation corresponding to the event includes a specific resource selected from at least the number of resources associated with each of the pairs of entities. In some implementations, generating the representation corresponding to the event includes: determining the type of the event based on information about the event associated with a particular pair of entities; and generating a second representation including a content item indicating the type of the selected information. In some implementations, at least one entity in the particular pair of entities is a business entity, the type of selected information corresponds to financial information, and the content item at least indicates the monetary characteristics of the business entity.

[0015] Other implementations of this aspect include corresponding computer systems, apparatuses, and computer programs recorded on one or more computer storage devices, each configured to perform the actions of these methods. A system of one or more computers may be configured to perform specific operations or actions by means of software, firmware, hardware, or combinations thereof installed on the system, which, upon operation, cause the system to perform these actions. One or more computer programs may be configured to perform specific operations or actions by means of instructions that, when executed by a data processing apparatus, cause the apparatus to perform these actions.

[0016] Another aspect of the subject matter described in this specification can be embodied in a method for populating an interest feed with electronic news article resources. The method is implemented by one or more processors and includes a method comprising: in response to determining that a threshold number of electronic news article resources all mention both a first entity and a second entity with a relevance of at least a threshold magnitude; determining an event that may include a specific activity involving both the first entity and the second entity; generating a representation corresponding to the event, the generation of the representation comprising: generating first content based on one or more of the electronic news article resources mentioning both the first entity and the second entity with a relevance of at least a threshold magnitude, and generating second content based on additional resources, wherein generating the second content based on the additional resources is based on determining that the additional resources relate to attributes of the specific activity included in the event, wherein the additional resources are resources other than the electronic news article resources mentioning both the first entity and the second entity with a relevance of at least a threshold magnitude, and wherein the attribute of the specific activity is an attribute other than the first entity and the second entity; identifying a user account including an interest list that includes the first entity but does not include the determined event; and in response to determining that the interest list includes the first entity and the event involves the first entity: providing the representation to a user device associated with the user account, wherein providing the representation causes the representation to be presented at the user device.

[0017] Another aspect of the subject matter described in this specification can be embodied in an electronic system. An electronic system includes: one or more processing devices; one or more non-transitory machine-readable storage devices for storing instructions executable by the one or more processing devices to cause the execution of operations including: in response to determining that a threshold number of electronic news article resources all mention both a first entity and a second entity with a relevance of at least a threshold magnitude; determining an event that may involve a specific activity relating to both the first entity and the second entity; generating a representation corresponding to the event, the generation of the representation including: generating first content based on one or more of the electronic news article resources mentioning both the first entity and the second entity with a relevance of at least a threshold magnitude, and generating second content based on additional resources, wherein generating the second content based on the additional resources is based on determining that the additional resources relate to attributes of the specific activity included in the event, wherein the additional resources are resources other than the electronic news article resources mentioning both the first entity and the second entity with a relevance of at least a threshold magnitude, and wherein the attribute of the specific activity is an attribute other than the first entity and the second entity; identifying a user account including an interest list that includes the first entity but does not include the determined event; in response to determining that the interest list includes the first entity and the event involves the first entity: providing the representation to a user device associated with the user account, wherein providing the representation causes the representation to be presented at the user device.

[0018] Another aspect of the subject matter described in this specification may be embodied in one or more non-transitory machine-readable storage devices. The one or more non-transitory machine-readable storage devices are used to store instructions that can be executed by one or more processing devices to cause the execution of operations, the operations including: identifying a first entity based on user search queries from multiple users; in response to determining that a threshold number of electronic news articles all mention both the first entity and the second entity with a relevance of at least a threshold magnitude: determining an event that may include a specific activity involving both the first entity and the second entity; generating a representation corresponding to the event, the generation of the representation including: generating first content based on one or more electronic news articles mentioning the first entity and the second entity with a relevance of at least a threshold magnitude, and generating second content based on additional resources, wherein generating the second content based on the additional resources is based on determining that the additional resources relate to attributes of the specific activity included in the event, wherein the additional resources are resources other than the electronic news articles mentioning the first entity and the second entity with a relevance of at least a threshold magnitude, and wherein the attribute of the specific activity is an attribute other than the first entity and the second entity; in response to identifying the first entity based on search queries from multiple users and determining that the event involves the first entity, determining that the list of interests includes the first entity and the event involves the first entity: providing the representation to a user device associated with a user account, wherein providing the representation causes the representation to be presented at the user device.

[0019] The subject matter described in this specification can be implemented in specific ways and can produce one or more of the following advantages. The described computing system can predict and identify specific electronic media content and resources that one or more users may be particularly interested in. Improved efficiency can be achieved in the computational processes of the computer system and user device by predicting and providing target media content / resources that are sufficiently matched to the user's interests or preferences. Computational efficiency can be improved by minimizing useless data processing steps that generate irrelevant media content that is not targeted at specific user preferences and can therefore be discarded by the user.

[0020] Details of one or more implementations of the subject matter described in this specification are set forth in the accompanying drawings and the following description. Other features, aspects, and advantages of the subject matter will become apparent from the description, drawings, and claims. Attached Figure Description

[0021] Figure 1 The illustration shows several interfaces related to interest feeding in the example computing system.

[0022] Figure 2 The diagram illustrates a system diagram of an example computational system used to populate interest feeds.

[0023] Figure 3 This is a flowchart of an example process for populating interest feeds.

[0024] Figure 4 This is a block diagram of a computing system that can be used in conjunction with the computer implementation methods described in this specification.

[0025] Similar labels and names in each figure indicate similar elements. Detailed Implementation

[0026] A computing system is described that receives various structured and unstructured data from diverse data sources, such as online article resources and other media content. The system is configured to annotate the received data and use the annotated data to automatically bundle and / or cluster specific content deemed relevant or of interest to at least one user. In this context, the described subject matter relates to enhancing user interest feeds to include targeted media content and resources that match the user's specific interests and preferences.

[0027] Figure 1 The illustrations depict multiple interfaces related to interest feeding in the example computing system. These interfaces include interfaces 102, 104, and 106. Each illustrated interface corresponds to an example graphical representation of a user interest feed that can be displayed on the example user device 108. Figure 1 As shown, user equipment 108 may correspond to a mobile smartphone device.

[0028] In some implementations, user device 108 can be one of a variety of computing devices, such as laptop / desktop computers, smart TVs, e-book / reader devices, digital streaming content devices, game consoles, smartwatches, wearable devices, tablets, or other related computing devices configured to execute software instructions and applications for providing target content to at least one user.

[0029] Interface 102 may be displayed on user device 108 and may include an example user interest feed 110. In some implementations, interest feed 110 displays a graphical representation corresponding to example media content, where the media content is, for example, an online article or other type of resource data. For example, interest feed 110 may include first content 112 and second content 114. As shown, first content 112 may indicate certain activities of an example business entity, namely company A, while second content 114 may indicate certain activities of another example business entity, namely company B.

[0030] First and second content 112, 114 of interest feed 110 can be generated according to one implementation. Content 112, 114 may include, from the perspective of a user of device 108, different entities that do not have obvious connections or associations. In some instances, potential connections, associations, or shared events involving the entities of interest feed 110 may actually exist. However, in one implementation, the system generating interest feed 110 does not detect relevance, nor does it identify available resource content describing relevance. Furthermore, the system does not infer or predict example second media content to be included in the user's interest feed based on the user's preference for a particular first media content.

[0031] Interfaces 104 and 106 each illustrate example graphical representations of interest feeds enhanced according to the described techniques. As shown, interfaces 104 and 106 can also be displayed on user device 108. Interface 104 includes a resource interest feed 116, while interface 106 includes an inferred interest feed 124. Interest feeds 116 and 124 each display graphical representations of example resource / media content, such as online news articles or other types of data, which are uniquely identified based on the user's specific interests or preferences.

[0032] Interest feed 116 can be populated to include first content 118 and second content 120. Unlike the content of interest feed 110, the first and second contents 118, 120 of interest feed 116 are generated by a computing system enhanced according to the described technology. The enhanced computing system can be configured to identify entities that match a user's preferences. For example, a user may have an interest or preference for news related to the business activities of a first identified entity, such as company A, and the enhanced computing system can be configured to use annotated resource data to detect a relevance or association between company A and a second identified entity, namely company B.

[0033] The enhanced system can also detect events involving entities and indicating connections between entities, and populate resource interest feed 116 to include sample e-articles or other digital resource content about the event and output it to the user via user device 108. In some implementations, the resource may be a URL for accessing online content describing or detailing events indicating connections or relevance between identified entities.

[0034] For example, as shown via first content 118, the resource could be a uniform resource locator (URL) for accessing an online article describing the epic merger between companies A and B. In some implementations, as shown via second content 120, interest feed 116 could be populated to include additional resource data, such as an example stock quote showing the respective stock prices 122 of companies A and B, and data on changes in the respective stock prices of each company's shares.

[0035] Similar to interest feed 116, inferred interest feed 124 may be populated to include first content 126 and second content 128 generated by a computing system enhanced according to the described techniques. The enhanced system may perform computations to identify or detect events to be included in inferred interest feed 124 based on inferences or predictions that match or are related to a user's specific interests or preferences.

[0036] In some implementations, events can also be inferred or predicted to be relevant to the user by the enhanced system based on their relevance or connection to another relevant event identified as of the user's interest. Thus, as described above, the interest feed 124 displays a graphical representation of resource / media content identified based on inferences determined using specific user preferences.

[0037] For example, as shown via the first content 126, information about the inferred event can be accessed via the URL of an online article describing the event as the recent acquisition of Company D by Company C. The event indicated by the URL can be predicted as of the user's interest based on their interest in Company A's business activities (e.g., a merger between Company A and Company B). Alternatively or additionally, the event can be predicted as of the user's interest based on its relevance to another relevant event identified as of the user's interest, such as an event generally related to business activities.

[0038] In some implementations, such as via second content 128, interest feed 124 may be populated to include additional resource data. Resources may be URLs / links to online content that describes or details specific data relating to the event detected in first content 126. For example, a resource may be a URL of a webpage describing facts and definitions of mergers and acquisitions (M&A) and other business concepts. Additional data may be determined by an enhanced system based on inferences or predictions made by the system regarding the event in first content 126. Inferences or predictions may be analytical calculations performed by the system, at least based on attributes of the event, to anticipate, identify, and select content that the user will be interested in.

[0039] For example, users may have preferences for news related to business activities (e.g., business activities involving at least Company A). Additionally, detected events may indicate that an M&A-type business activity is occurring or will occur between Company A and at least one other company. Thus, the enhanced system allows the inferred interest feed 124 to be populated with predictions that the user might be interested in learning more about business definitions or other financial concepts, taking into account the attributes of the business activities involved in Company A (e.g., mergers).

[0040] Figure 2 The diagram illustrates a system diagram of an example computational system for populating interest feeds. System 200 generally includes components for at least generating the above-mentioned references. Figure 1 The example computing server 109 is described in interfaces 104 and 106. In some implementations, the computing capabilities of server 109 may be performed in the example cloud-based computing system or environment.

[0041] System 200 corresponds to the reference above. Figure 1 The described enhanced computing system is used to generate interfaces 104 and 106 displayed using user device 108. Therefore, system 200 can be configured to accurately detect potential connections, associations, or shared events involving two or more entities based on user preferences or interests. System 200 can also be configured to identify available resources or media content describing such correlations, such as online articles about merger events, and provide them to the user.

[0042] System 200 can also identify additional content related to events indicating relevance and provide it to the user. For example, additional content could be the stock prices of the two merging companies. System 200 can also be configured to identify or detect events and related data based on inferences or predictions matching specific user preferences. In some implementations, the connection between an event and another relevant event identified by System 200 as of user interest can be used to infer events and related data.

[0043] refer to Figure 2 Server 109 generally includes an event detection module 202, a resource interest feeding module 204, and an inferred interest feeding module 206. In some implementations, modules 202, 204, and 206 are each included collectively or individually in server 109 or are accessible by server 109. Furthermore, the functions and computational processes described for modules 202, 204, and 206 can be enabled by computational logic or programming instructions executable by the processor and memory associated with server 109.

[0044] As used in this specification, the term "module" is intended to include—but is not limited to—one or more computers / computing devices configured to execute software programs, including program code that causes one or more processing units of the computing device(s) to perform one or more functions. The term "computer" is intended to include any data processing or computing device / system, such as a desktop computer, laptop computer, mainframe computer, personal digital assistant, server, handheld device, or any other device capable of processing data.

[0045] Server 109 may include one or more processors, memory, and data storage devices that generally form one or more computing systems that make up server 109. The processors of the computing systems process instructions for execution by server 109, including instructions stored in memory or on storage devices to display graphical information of a graphical user interface (GUI) via an example display of user device 108. Execution of the stored instructions may cause one or more of the actions described herein to be performed by server 109 or user device 108.

[0046] In other implementations, multiple processors, as well as multiple memories and different types of memory, may be used as appropriate. For example, server 109 may be connected to multiple other computing devices, each of which (e.g., a server array, a group of servers, a module, or a multiprocessor system) performs some part of the actions or operations associated with the various processes or logical flows described in this specification.

[0047] Refer again Figure 2 The event detection module 202 includes a resource annotator 208, an entity relevance detector 210, and an event detector 212. Module 202 also includes multiple resources 216, user interest profiles 218, and filtering logic 220. The annotator 208 can be used by the system 200 to annotate each of the multiple resources 216 or media content received by the server 109. For example, the annotator 208 can annotate each received resource by scanning or analyzing entity or content data about the resource.

[0048] In some implementations, to annotate resource 216, an annotator 208 analyzes signal data associated with individual entities or other content included in each resource 216. Based on this signal analysis, the annotator 208 can identify and / or extract one or more entities included in the content data for or about each resource 216. In some implementations, the extracted entities are stored in the memory of server 109. The stored entities can then be accessed and used by modules of server 109 to perform calculations for populating resource interest feed 116 or inferring interest feed 124.

[0049] As described above, the entities in Resource 216 can span multiple information types. For example, the entities in Resource 216 can include individuals, government and political entities, sports and entertainment entities, or business, scientific and academic entities. As described above, in some instances, two or more entities may be related or connected, and example events involving two entities may be included in at least one resource item in Resource 216.

[0050] The entity relevance detector 210 may receive annotated resource items and clusters of identified entities from the annotator 208. For example, the relevance detector 210 may receive a cluster of multiple electronic articles, including associated names or identifiers of individual entities included in the articles. In some implementations, for each resource in the cluster of multiple resources, the relevance detector 210 may analyze the resource and, based on the analysis, identify one or more entities, identify relevances between resources, or identify relevances between entities associated with various resources.

[0051] For example, relevance detector 210 can analyze resources to detect or determine data similarity or related data between electronic text or digital images of multiple resources or media content. In some implementations, relevance detector 210 uses the relative position of specific text or image content of each resource, such as the title / heading of an article, to determine resource or entity relevance between the various resources.

[0052] In response to determining data similarity, the correlation detector 210 may identify correlations between resources or correlations between entities associated with each resource, based at least on the number of detected data similarities. For example, the correlation detector 210 may determine the number of resources with detected data similarities that can indicate correlations or associations between two or more entities.

[0053] The correlation detector 210 can then identify two or more entities that are related or associated based on the determined number of resources. In some instances, the correlation detector 210 identifies two or more entities based on the determined number of resources exceeding a threshold.

[0054] In some implementations, the relevance detector 210 uses a cluster of annotated resources to identify at least one entity that matches or is associated with the user's interests or preferences. The relevance detector 210 may then identify at least one other entity that is associated with or related to the at least one entity and / or also matches the user's interests or preferences. In some instances, the relevance detector 210 analyzes the cluster of resources to identify entity pairings based on the user's preferences and / or based on resource or entity relevance determined using detected data similarity.

[0055] For example, relevance detector 210 can use the entity names of companies or people included in various electronic articles in the cluster to identify at least one entity, Company A, that matches the user's business activity interests. Relevance detector 210 can then determine at least one other entity, such as Company B, that is associated with or relevant to Company A based on the user's business activity interests / preferences.

[0056] In some instances, entities Company A and Company B can be identified based on a first electronic article whose title contains Company A and a second electronic article whose title contains Company B. Furthermore, the relevance between Company A and Company B can be determined, for example, based on one or more detected similarities between the text / image content of the first article about Company A and the second article about Company B.

[0057] As described in more detail below, in addition to the relevance detector 210, server 109 and system 200 may use other computational logic of module 202 to determine the first entity and at least one other entity associated with the first entity. For example, module 202 may be used to track at least one trending topic. Trending topics may be determined based on data associated with multiple user search checks, such as unstructured data. Module 202 may then determine entity relevance based on the number of resources that describe or include entities or other content related to one or more trending topics (described below).

[0058] In some instances, the relevance detector 210 may be configured to detect or identify multiple entities, rather than pairs of entities, or to identify multiple pairs of entities. The relevance detector 210 may also be configured to detect multiple sets of entities, wherein the sets of entities include two or more entities.

[0059] The relevance detector 210 can access the user interest profile 218 to obtain or determine one or more preferences of the user used to identify entities. In some implementations, the user profile 218 may be associated with an example user account and may be populated with user preferences based on the user's interests / preferences defined through the user account profile. Additionally or alternatively, the user profile 218 may be populated with user preferences based on interests or preferences associated with the user's online browsing activity or based on various other online resources about the user or various account profiles of the user.

[0060] In some implementations, the relevance detector 210 generates individual resource and entity scores that estimate the magnitude of the relevance of the electronic resource 216 and its entities to the specific interests of various users. For example, an article resource may include the article's title and the body of the article describing its substance. For users interested in the business activities of various companies, an electronic article that includes Company A in the title and includes multiple mentions of Company A throughout the body may receive a relatively high resource-entity score, such as 0.9.

[0061] However, e-articles that include Restaurant X in the title but only a few mentions of Company A throughout the body may receive a relatively low resource-entity score, such as 0.2. The relevance detector 210 can use resource-entity scores to select a subset of resources and corresponding entities from which relevant entities and one or more events can be identified or detected. For example, the relevance detector 210 may select a subset of e-articles with individual resource-entity scores exceeding a threshold score (e.g., 0.7) and the corresponding individual or company names of these articles.

[0062] Generally, the relevance detector 210 can use one or more of the following: data similarity between detected resources 216, entity data related to trending topics, or user interest data, to identify two or more relevant entities associated with a subset of resources. For at least for each pair of identified entities among the two or more relevant entities, the relevance detector 210 can determine the number of resources associated with each entity in that pair. From the determined number of resources, the relevance detector 210 can select a subset of resources and corresponding entities, from which one or more events can be identified or detected, and each of these one or more events includes the identified relevant entity.

[0063] Event detector 212 receives a subset of resources and corresponding entities, such as entity pairs (or sets) identified and selected by correlation detector 210. The subset of resources can be analyzed by event detector 212 to detect one or more events. For example, event detector 212 can execute software instructions to analyze individual resources and entities with obvious associations or correlations. Based on this analysis, event detector 212 can detect or identify one or more events described by a resource and involving at least one entity, at least one entity pair, or at least one set of entities.

[0064] For example, as described in more detail below, a subset of resources may include electronic documents relating to the business activities of the paired entities—Company A and Company B. Referring to the first content 118 of interface 104, event detector 212 can detect events related to the merger of Company A and Company B and relating to epic merger events in the business industry. Events may be associated with or described in a specific resource within the subset of resources received from relevance detector 212.

[0065] Event detector 212 can detect one or more events, including pairs or sets of related entities, based on existing associations between one or more entities in a subset of resources. In some implementations, event detector 212 determines the occurrence of an event associated with a particular pair of entities based on the number of resources identified. For example, event detector 212 can identify multiple candidate events associated with a specific pair of entities based on a subset of resources, and then determine the occurrence of the event from among the multiple candidate events.

[0066] Event detector 212 can identify at least one event from a plurality of candidate events, wherein identifying an event includes selecting one or more resources associated with the event. In some implementations, the selected resource is an electronic article describing the event or including text or image content about the event. As described below, system 200 can provide one or more of the selected resources to a user device so that the resources can be viewed by a user via the user device's display.

[0067] The filtering logic 220 can be used by the event detector 212 (or the relevance detector 210) to identify article resources or other media content that indicate an event matching at least one of the user's interests or preferences. For example, see reference Figure 1 The individual electronic articles of the first content 112 and the second content 114 may be included in a subset of the resources identified by the relevance detector 210 and received by the event detector 212. Therefore, Company A and Company B may be corresponding entities forming a pair of entities identified by the relevance detector 210.

[0068] In addition, such as Figure 1 As shown, the article in the first content 112 may be related to an event involving Company A, which is related to business activities, namely the acquisition (or merger) of another company. The article in the second content 114 may be related to an event involving Company B, which is only loosely or slightly related to business activities. Therefore, based on user preferences or interests regarding articles about business activities, the resource-entity score of the article resource in the first content 112 may be slightly higher than the resource-entity score of the article resource in the second content 114.

[0069] In some implementations, event detector 212 identifies one or more entities associated with at least one article resource to be provided to users. Event detector 212 (or relevance detector 210) may use a collaborative filtering algorithm, such as filtering logic 220, to determine one or more interests of a first user based at least on the interests of one or more second users. Event detector 212 may then use the determined interests(s) of the first user(s) to filter out or remove at least one event from candidate events.

[0070] In some implementations, events can be filtered out or removed from candidate events when the event includes entities that do not match the interests of one or more of the first user, or involves other content or activities that do not match the interests of one or more of the first user. In some instances, filtering events from candidate events includes removing at least one article resource from a subset of resources, wherein the removed article resource is associated with the event.

[0071] Filtering logic 220 includes trending content 222 that identifies resource content associated with multiple trending topics. In some instances, trending content 222 may correspond to computational logic used to track, analyze, and identify multiple resources related to trending content / topics. As mentioned above, trending content 222 may be determined based on data associated with multiple user search queries received and processed by an example search system (e.g., Google Search). In some implementations, trending content 222 is determined by analyzing multiple digital news and social media content generated by one or more web-based information systems.

[0072] Event detector 212 may use trend content 222 to identify or determine event and article resources that describe or include entities or other content corresponding to one or more trending topics. In some implementations, event detector 212 uses filtering logic 220 and trend content 222 to determine event and article resources that also match trending topics that match the user's interests or preferences.

[0073] Event detector 212 can then use the identified events corresponding to the trending topic to filter or remove at least one event from the candidate events. In some implementations, events can be filtered or removed from the candidate events when they include entities not identified in the trending content 222 and / or not matching the user's interests(s) or involve other content not identified in the trending content 222 and / or not matching the user's interests(s).

[0074] Event detector 212 can use knowledge graph 214 to detect one or more events involving entities. In some implementations, knowledge graph 214 is used and / or analyzed to determine activities and other content that can indicate events. In some instances, sets or pairs of entities can be identified as having particularly strong associations or relevance based on additional relevance data available via knowledge graph 214.

[0075] Knowledge graph 214 can be used to select one or more events from candidates received at modules 204 and 206. For example, event detector 212 can access knowledge graph 214 to obtain data indicating events involving at least two entities, where the events and entities match a user's specific interests, such as business or engineering interests. In response to obtaining data from knowledge graph 214, at least one article resource corresponding to a specific candidate event can be provided to modules 204 and / or 206 based on the correlation between the obtained data and the specific candidate event.

[0076] Knowledge Graph 214 can be represented by any of a variety of convenient physical data structures. For example, Knowledge Graph 214 can be represented by triples, where each represents two entities in sequence and a relation from the first entity to the second entity; for example, [Alpha, Beta, is parent] or [Alpha, is parent, Beta] are alternative ways of representing the same fact. Each entity and each relation can be included in multiple triples and will generally be included in multiple triples.

[0077] Alternatively, each entity can be stored, for example, once as a node, as a record, or as an object, and linked to all the relationships that the entity has and all other entities related to it via a linked list data structure. More specifically, knowledge graph 214 can be stored as an adjacency list, where adjacency information includes relationship information. It is generally advantageous to use unique identifiers to represent each distinct entity and each distinct relationship.

[0078] The entities represented in Knowledge Graph 214 do not need to be tangible things or specific people. Entities can include specific people, places, things, works of art, concepts, events, or other types of entities. Thus, Knowledge Graph 214 can include data defining relationships between people, such as co-stars in a movie; data defining relationships between people and things, such as a specific singer recording a specific song; data defining relationships between places and things, such as a specific type of wine coming from a specific geographical location; data defining relationships between people and places, such as a specific person being born in a specific city; and other kinds of relationships between entities.

[0079] In some implementations, each node has a type based on the kind of entity it represents; and these types may each have a schema specifying the kind of data that can be maintained for the entity represented by a node of that type and how that data should be stored. Thus, for example, a node representing a person type may have a schema defining fields for information such as birthday, place of birth, etc. This information may be represented by fields in a type-specific data structure, or by triples that look like node-relationship-node triples, such as [person identifier, born, date], or by any other convenient predefined method. Alternatively, some or all of the information specified by the type schema may be represented by links to nodes in knowledge graph 214; for example, [one person identifier, child, another person identifier], where the other person identifier is a node in the graph.

[0080] The resource interest feed module 204 includes a narrative generator 224 and an interest feed filler 226. In some implementations, the functions / features of the described narrative generator 224 and interest feed filler 226 correspond to the computational process of module 204 enabled by executable software instructions or computational logic. The resource output 228 generally corresponds to a graphical representation generated to be displayed to a user via an example user device.

[0081] System 200 may provide resource output 228 as a resource representation to be output to a user via a display of user equipment 108. For example, as shown above, event detector 212 selects at least one event from a plurality of candidate events and an electronic article resource associated with that event. In some instances, module 204 receives a plurality of selected events and article resources corresponding to each selected event. System 200 causes data for accessing the article resource associated with each selected event to be included in interest feeds 116 or 124. Interest feeds 116 and 124 may include data representing resource output 228, wherein the interest feed is provided to user equipment 108 for output at that device.

[0082] In some implementations, system 200 generates a representation, such as resource output 228, including web links to resources for each selected event, and provides this representation to be output at user device 108. For example, system 200 may generate resource output 228 based on resources that describe the selected event and are associated with each entity in a pair of entities. System 200 may then use interest feed populator 226 to populate interest feed 116 or 124 to include resource output 228 or data related to the content of resource output 228.

[0083] System 200 may use narrative generator 224 to generate a narrative or summary about each resource corresponding to a specific event. For example, narrative generator 224 may include software instructions for an example data analyzer program for scanning or analyzing the electronic text, image, or media content of a resource. In response to analyzing the content of a resource using the data analyzer, narrative generator 224 may generate a data structure including text, image, or media content that provides a narrative or summary of the resource.

[0084] System 200 allows each resource summary and web link for each resource in an event to be included in the resource output 228 or in... Figure 1 In specific interest feeds. In some implementations, narrative generator 224 can generate text, images, or other data indicating a specific event and why an article resource for that specific event is chosen to be output to the user. In some instances, narrative generator 224 can access user interest profile 218 to generate data that informs the user why specific event / resource content is being presented in the user's interest feed.

[0085] For example, the reason could be that resources for a specific entity or event are provided to the user based on the user's interest in a specific business activity or the user's interest in a specific entity. In some implementations, system 200 includes one or more of the following in resource output 228 or includes... Figure 1 In the specific interest feed: i) article resources; ii) resource summaries; iii) web links to resources used to access the event; and iv) the reason for providing a specific web link or a specific resource describing a specific event.

[0086] The inference interest feed module 206 includes an event type recognizer 230 and an inference generator 232. In some implementations, the functions / characteristics of the described event type recognizer 230 and inference generator 232 correspond to the computation process of module 206 enabled by executable software instructions or computational logic. Similar to resource output 228, inference output 234 generally corresponds to a graphical representation generated to be displayed to the user via the display of user equipment 108.

[0087] System 200 may provide inference output 234 as an inference representation to be output to a user. In some implementations, inference output 234 corresponds to a graphical representation of inference data related to a specific event, where the inference data is, for example, a webpage, a URL for accessing a webpage, or an electronic article. Inference output 234 may include data related to a specific user interest topic or specific event selected from a group of candidate events identified by event detector 212.

[0088] As shown above, system 200 may use inference generator 232 to identify or predict events to be included in inference interest feed 124 based on inferences or predictions that match or are related to a user's specific interests or preferences. In some implementations, system 200 uses inference generator 232 to populate interest feed 124 with additional resource data. This additional resource data may be web links to online content related to or detailing the article resources included in resource output 228.

[0089] System 200 may use event type recognizer 230 to identify the type of a specific event. For example, event type recognizer 230 may include software instructions for analyzing electronic text, images, or media content of a resource describing the event. In response to analyzing the content of the resource, event type recognizer 230 may determine the type of the event based on data / information about events associated with a specific entity pair. System 200 may then use event type recognizer 230 to generate a representation indicating the type of event.

[0090] For example, event type identifier 230 can analyze the data content of resources describing the event. Based on the analyzed data content, event type identifier 230 can determine whether the event type is a business event, an engineering event, a global news event, an entertainment or social media event, or any other type of event tag that summarizes the event.

[0091] Event type recognizer 230 can then use the determined event type to generate an example digital banner, such as a representation, that includes text or image data indicating the event type. In response to event type recognizer 230 determining the event type, system 200 can cause the event type to be included in the inference output 234 and provided to the user via interest feed 124.

[0092] Figure 3 This is a flowchart of an example process for populating an interest feed. Process 300 can be implemented using system 200 described above. Therefore, the description of process 300 may refer to one or more of the modules or computing devices described above in system 200. In some implementations, the operation of the described process 300 is enabled by computational logic or software instructions executable by the processor and memory of an example electronic device, such as server 109 or user device 108 described above.

[0093] In block 302 of process 300, system 200 receives multiple resources, each of which may include electronic media content, such as text data, image data, and other media data / content. For example, system 200 may receive multiple electronic articles including written descriptions and digital images related to various topics. The received articles may correspond to article citations indicated via interfaces 102, 104, and 106.

[0094] In box 304, for each of the multiple resources, system 200 can identify the entities associated with that resource. For example, an article may describe entities such as Company A and Company B, and system 200 can analyze the article's data content to parse, copy, or extract entity names and other textual content, such as person names and company names.

[0095] In some implementations, system 200 analyzes real-time received user interest / preference data, or stored interest data, to determine if a user is interested in engineering, physics, or other scientific topics. The system can then identify entities associated with the resource based on the determined user interests. For example, an article might include a written description of a new commentary on Albert Einstein's quantum mechanics equations interpreted by a popular engineer X from company Y. Therefore, the identified entities could include "engineer X," "company Y," or "Albert Einstein."

[0096] In block 306 of process 300, for each pair of identified entities: system 200 determines the number of resources associated with each entity in the pair. The number of resources may form a subset of online articles describing various data items related to engineering or physics.

[0097] In box 308, system 200 determines the occurrence of an event associated with a particular pair of entities based on the determined number of resources. Determining the occurrence of an event associated with an entity may include identifying or generating a subset of candidate events and selecting at least one article describing the event.

[0098] For example, system 200 may identify a subset of online articles describing quantum mechanics, Albert Einstein, and / or Company X, such as 5-10 articles. System 200 may generate a subset of candidate events related to physics and / or these entities. The system may then use filtering logic 220 to determine if the user has a particular interest in general relativity rather than quantum mechanics.

[0099] System 200 can then determine the occurrence of an event by selecting online article resources from a subset of new commentaries describing Einstein's views on general relativity. The selected articles may describe physics conference events on the theory of relativity and symposium materials for engineers used in those events. Alternatively, another selected article may describe upcoming publications of new physics papers describing engineers' entirely new commentaries on the theory of relativity.

[0100] In block 310 of process 300, system 200 generates a representation corresponding to the event. This representation can be generated based on online articles that associate each entity in the pair of entities. In block 312 of process 300, system 200 provides the representation corresponding to the event to user device 108.

[0101] System 200 may use each of modules 204 and 206 to provide an example interest feed configured to be readily apparent, which may include the user's favorite content topics and interests, and include media content related to trending global and local news and other events. For example, the interest feed may include web links to articles about physics conference events and upcoming publications.

[0102] In some implementations, system 200 uses one or more of these modules to anticipate or predict content that will be of interest and importance to the user. For example, system 200 may provide one or more web links to online resources describing scientific data about general relativity based on predicted user interests. System 200 uses interest feeds to provide web links and additional annotated content indicating a summary of the data, the type of content (e.g., physics / science), and the reason for providing the content.

[0103] Figure 4 These are block diagrams of computing devices 400 and 450, which can be used as clients or as servers or multiple servers to implement the systems and methods described herein. Computing device 400 is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframes, and other suitable computers. Computing device 450 is intended to represent various forms of mobile devices, such as personal digital assistants, cellular phones, smartphones, smartwatches, head-mounted devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are intended to be illustrative only and are not intended to limit the implementations described and / or claimed herein.

[0104] Computing device 400 includes a processor 402, a memory 404, a storage device 406, a high-speed interface 408 connected to the memory 404 and a high-speed expansion port 410, and a low-speed interface 412 connected to a low-speed bus 414 and a storage device 416. Each of components 402, 404, 406, 408, 410, and 412 is interconnected using various buses and may be mounted on a common motherboard or otherwise, as appropriate. Processor 402 can process instructions for execution within computing device 400, including instructions stored in memory 404 or on storage device 406 for displaying graphical information for a GUI on an external input / output device, such as a display 416 coupled to high-speed interface 408. In other implementations, multiple processors and / or multiple buses, as well as multiple memories and multiple types of memories, may be used as appropriate. Additionally, multiple computing devices 400 may be connected, with each device providing a portion of the necessary operation (e.g., as a server array, a group of blade servers, or a multiprocessor system).

[0105] Memory 404 stores information within computing device 400. In one implementation, memory 404 is a computer-readable medium. In one implementation, memory 404 is one or more volatile memory cells. In another implementation, memory 404 is one or more non-volatile memory cells.

[0106] Storage device 406 provides large-capacity storage for computing device 400. In one implementation, storage device 406 is a computer-readable medium. In various implementations, storage device 406 may be a hard disk drive, optical disk drive, magnetic tape drive, flash memory or other similar solid-state storage device, or an array of devices, including devices in a storage area network or other configuration. In one implementation, a computer program product is tangibly embodied in an information carrier. The computer program product contains instructions that, when executed, perform one or more methods, such as those described above. The information carrier is a computer-readable or machine-readable medium, such as memory 404, storage device 406, or memory on processor 402.

[0107] High-speed controller 408 manages bandwidth-intensive operations for computing device 400, while low-speed controller 412 manages lower bandwidth-intensive operations. This allocation of functions is merely exemplary. In one implementation, high-speed controller 408 is coupled to memory 404, display 416 (e.g., via a graphics processor or accelerator), and high-speed expansion port 410, which accepts various expansion cards (not shown). In this implementation, low-speed controller 412 is coupled to storage device 406 and low-speed expansion port 414. The low-speed expansion port, which may include various communication ports (e.g., USB, Bluetooth, Ethernet, Wireless Ethernet), may be coupled to one or more input / output devices, such as keyboards, pointing devices, scanners, or networking devices such as switches or routers (e.g., via network adapters).

[0108] The computing device 400 can be implemented in various different forms, as shown in the figure. For example, it can be implemented as a standard server 420, or multiple times in a group of such servers. It can also be implemented as part of a rack server system 424. Furthermore, it can be implemented in a personal computer such as a laptop computer 422. Alternatively, components from the computing device 400 can be combined with other components in a mobile device (not shown) (e.g., device 450). Each such device can contain one or more of the computing devices 400, 450, and the entire system can consist of multiple computing devices 400, 450 communicating with each other.

[0109] The computing device 450 includes a processor 452, a memory 464, input / output devices such as a display 454, a communication interface 466 and a transceiver 468, and other components. The device 450 may also include a storage device, such as a micro hard disk or other device, to provide additional storage. Each of the components 450, 452, 464, 454, 466, and 468 is interconnected using various buses, and several of the components may be mounted on a common motherboard or otherwise, as appropriate.

[0110] Processor 452 can process instructions for execution within computing device 450, including instructions stored in memory 464. The processor may also include separate analog and digital processors. The processor can support, for example, coordination with other components of device 450, such as control of the user interface, applications running on device 450, and wireless communications performed by device 450.

[0111] Processor 452 can communicate with the user via control interface 458 and display interface 456 coupled to display 454. Display 454 may be, for example, a TFT LCD display or an OLED display, or other suitable display technology. Display interface 456 may include appropriate circuitry for driving display 454 to present graphics and other information to the user. Control interface 458 can receive commands from the user and translate them to submit to processor 452. Additionally, an external interface 462 may be provided to communicate with processor 452 to enable near-field communication between device 450 and other devices. External interface 462 may, for example, support wired communication (e.g., via a docking process) or wireless communication (e.g., via Bluetooth or other such technologies).

[0112] Memory 464 stores information within computing device 450. In one implementation, memory 464 is a computer-readable medium. In another implementation, memory 464 is one or more volatile memory cells. In yet another implementation, memory 464 is one or more non-volatile memory cells. Extended memory 474 may also be provided, and it can be connected to device 450 via an extended interface 472, which may include, for example, a SIMM card interface. This extended memory 474 can provide additional storage space for device 450, or it may store applications or other information for device 450. Specifically, extended memory 474 may include instructions for performing or supplementing the processes described above, and may also include security information. Thus, for example, extended memory 474 may be provided as a security module for device 450 and may be programmed with instructions allowing secure use of device 450. Furthermore, security applications may be provided via a SIMM card along with additional information, such as placing identification information on the SIMM card in an unbreakable manner.

[0113] The memory may include, for example, flash memory and / or MRAM memory, as described below. In one implementation, the computer program product is tangibly embodied in an information carrier. The computer program product contains instructions that, when executed, perform one or more methods, such as those described above. The information carrier is a computer-readable or machine-readable medium, such as memory 464, extended memory 474, or memory on processor 452.

[0114] Device 450 can communicate wirelessly via communication interface 466, which may include digital signal processing circuitry if necessary. Communication interface 466 can support communication under various modes or protocols, such as GSM voice calls, SMS, EMS or MMS messaging, CDMA, TDMA, PDC, WCDMA, CDMA2000, or GPRS, etc. This communication can occur, for example, via radio frequency transceiver 468. Furthermore, short-range communication can occur, for example, using Bluetooth, WiFi, or other such transceivers (not shown). Additionally, GPS receiver module 470 can provide additional wireless data to device 450, which can be used as appropriate by applications running on device 450.

[0115] Device 450 may also utilize audio codec 460 for audible communication, which receives spoken information from a user and converts it into usable digital information. Audio codec 460 may similarly generate audible sounds for the user, for example, through a speaker, such as the speaker in the handset of device 450. Such sounds may include sounds from voice telephone calls, recorded sounds (e.g., voice messages, music files, etc.), and sounds generated by applications operating on device 450.

[0116] The computing device 450 can be implemented in a variety of different forms, as shown in the figure. For example, it can be implemented as a cellular phone 480. It can also be implemented as part of a smartphone 482, a personal digital assistant, or other similar mobile device.

[0117] Various implementations of the systems and techniques described herein can be implemented in digital electronic circuits, integrated circuits, specially designed ASICs, computer hardware, firmware, software, and / or combinations thereof. These various implementations may include implementations in one or more computer programs executable and / or interpretable on a programmable system, which includes at least one programmable processor, which may be dedicated or general-purpose, coupled to receive and send data and instructions from and to a storage system, at least one input device, and at least one output device.

[0118] These computer programs—also known as programs, software, software applications, or code—comprise machine instructions for programmable processors and can be implemented using high-level procedural and / or object-oriented programming languages ​​and / or assembly / machine languages. Programs may be stored as a portion of a file containing other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program of interest, or in multiple coordinated files (e.g., a file storing one or more modules, subroutines, or code sections). Computer programs may be deployed to execute on one or more computers located in one location or distributed across multiple locations and interconnected by a communication network.

[0119] When used herein, the terms "machine-readable medium" and "computer-readable medium" refer to any computer program product, apparatus, and / or device used to provide machine instructions and / or data to a programmable processor, such as a magnetic disk, optical disk, memory, or programmable logic device (PLD), including machine-readable media that receive machine instructions in the form of machine-readable signals. The term "machine-readable signal" refers to any signal used to provide machine instructions and / or data to a programmable processor.

[0120] To support interaction with the user, the systems and techniques described herein can be implemented on a computer having a display device for displaying information to the user, such as a CRT (cathode ray tube) or LCD (liquid crystal display) monitor, and a keyboard and pointing device, such as a mouse or trackball, that the user can use to provide input to the computer. Other types of devices may also be used to support interaction with the user; for example, feedback provided to the user can be any form of sensory feedback, such as visual, auditory, or tactile feedback; and input from the user can be received in any form, including sound, voice, or tactile input.

[0121] The systems and technologies described herein can be implemented in a computing system that includes back-end components, such as a data server, or middleware components, such as an application server, or front-end components, such as a client computer having a graphical user interface or web browser through which a user can interact with the implementation of the systems and technologies described herein, or any combination of such back-end, middleware, or front-end components. The components of the system can be interconnected by digital data communications (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0122] A computing system may include clients and servers. Clients and servers are generally geographically separated and typically interact through communication networks. The client-server relationship arises from the fact that computer programs run on various computers and have client-server relationships with each other.

[0123] Regarding the above description, furthermore, users may be provided with controls to allow them to make choices regarding whether and when the system, program, or feature described herein may enable the collection of user information (e.g., information about the user's social networks, social actions or activities, occupation, user preferences, or the user's current location), and whether the user is sent content or communications from the server. Furthermore, before storing or using certain data, this data may be processed in one or more ways to remove personally identifiable information.

[0124] For example, in some embodiments, a user's identity may be processed such that personally identifiable information cannot be determined for the user, or, where location information is available, the user's geographic location may be generalized (e.g., to the city, zip code, or state level) so that a specific location for the user cannot be determined. Thus, the user has control over what information is collected about them, how that information is used, and what information is provided to them.

[0125] Several embodiments have been described. However, it will be understood that various modifications can be made without departing from the spirit and scope of the invention. Various forms of the processes shown above can be used, wherein steps are reordered, added, or deleted. Furthermore, while several applications of the payment system and method have been described, it should be recognized that many other applications are contemplated. Therefore, other embodiments are within the scope of the appended claims.

[0126] Specific embodiments of the subject matter have been described. Other embodiments are within the scope of the appended claims. For example, the actions recited in the claims may be performed in a different order while still achieving the desired result. As an example, the processes depicted in the drawings do not necessarily require the specific order or sequence shown to achieve the desired result. In some cases, multitasking and parallel processing may be advantageous.

Claims

1. A method for populating an interest feed with electronic news article resources, the method being implemented by one or more processors and comprising: In response to determining a threshold number of electronic news articles that all mention both a first entity and a second entity, wherein the magnitude of the correlation between the first entity and the second entity is at least a threshold magnitude, an event that may involve a specific activity is determined, the specific activity involving both the first entity and the second entity. Generate a representation corresponding to the event. The generated representation includes: First content is generated based on one or more electronic news article resources that mention both the first entity and the second entity, and Generating second content based on additional resources, wherein generating second content based on additional resources is based on determining that the additional resources are related to the attributes of specific activities included in the event. The additional resources are resources other than electronic news articles that mention both the first entity and the second entity, and Among them, the attributes of a specific activity are those other than the first entity and the second entity; Identify user accounts that include a list of interests, which includes the first entity but excludes the identified events; and In response to determining that the list of interests includes the first entity and the event involves the first entity: The representation is provided to the user device associated with the user account, wherein providing the representation causes the representation to be presented at the user device.

2. The method according to claim 1, wherein determining that the electronic news article resources all mention both the first entity and the second entity comprises: For each electronic news article resource, generate a first resource entity score corresponding to the first entity and a second resource entity score corresponding to the second entity, and... Determine whether the scores of the corresponding first resource entity and the corresponding second resource entity satisfy the threshold.

3. The method according to claim 1, wherein determining that the electronic news article resources all mention both the first entity and the second entity comprises: Identify supersets of electronic news article resources that mention both the first and second entities; For each electronic news article resource in the superset, generate the first resource entity score corresponding to the first entity and the second resource entity score corresponding to the second entity. Based on the corresponding first resource entity score and the corresponding second resource entity score, electronic news articles that mention both the first and second entities are selected from the superset. Both mention that the electronic news article resources of the first entity and the second entity are subsets of a superset.

4. The method of claim 1, wherein determining that the event may occur comprises: Identify multiple candidate events that are associated with both the first entity and the second entity; A filtering algorithm is used to generate a subset of candidate events associated with e-news article resources that mention both the first entity and the second entity. as well as The possibility of an event is determined based on at least one event included in a subset of candidate events.

5. The method of claim 1, wherein generating a representation corresponding to the event further comprises: Based on one or more electronic news article resources that mention both the first entity and the second entity, determine the type of event that may have occurred. The representation corresponding to an event is generated based on the type of the event.

6. The method of claim 1, wherein the representation corresponding to the event includes content describing the possible events.

7. The method of claim 1, wherein generating the second content based on additional resources comprises: The relevance between the first entity and the second entity is determined based on a subset of electronic news article resources that mention both the first entity and the second entity. Identify the attributes of specific activities included in possible events; Based on additional resources describing the correlation between the first and second entities, additional resources are selected for generating the second content; and Secondary content is generated based on the attributes of specific activities included in events related to additional resources.

8. The method of claim 7, wherein the attributes of the specific activity included in the event include one or more of the following: Activity types corresponding to specific activities, Industries associated with specific activities, The connection or correlation between this event and previously identified events, including specific activities, or The correlation between events and lists of interests of different user accounts associated with different users.

9. An electronic system comprising: One or more processing devices; One or more non-transitory machine-readable storage devices are used to store instructions executable by one or more processing devices to cause the execution of operations including: In response to determining a threshold number of electronic news articles that all mention both a first entity and a second entity, wherein the magnitude of the correlation between the first entity and the second entity is at least a threshold magnitude, an event that may involve a specific activity is determined, wherein the specific activity involves both the first entity and the second entity. Generate a representation corresponding to the event. The generated representation includes: First content is generated based on one or more electronic news article resources that mention both the first entity and the second entity, and Generating second content based on additional resources, wherein generating second content based on additional resources is based on determining that the additional resources are related to the attributes of specific activities included in the event. The additional resources are resources other than electronic news articles that mention both the first entity and the second entity, and Among them, the attributes of a specific activity are those other than the first entity and the second entity; Identify user accounts that include a list of interests, which includes the first entity but excludes the identified events; In response to determining that the list of interests includes a first entity and the event involves the first entity: The representation is provided to the user device associated with the user account, wherein providing the representation causes the representation to be presented at the user device.

10. The system of claim 9, wherein determining that the electronic news article resources all mention both the first entity and the second entity comprises: For each electronic news article resource, generate a first resource entity score corresponding to the first entity and a second resource entity score corresponding to the second entity, and... Determine whether the scores of the corresponding first resource entity and the corresponding second resource entity satisfy the threshold.

11. The system of claim 9, wherein determining that the electronic news article resources all mention both the first entity and the second entity comprises: Identify a superset of electronic news article resources that mention both the first entity and the second entity; For each electronic news article resource in the superset, generate a first resource entity score for the first entity and a second resource entity score for the second entity. Based on the corresponding first resource entity score and the corresponding second resource entity score, electronic news articles that mention both the first and second entities are selected from the superset. Both mention that the electronic news article resources of the first entity and the second entity are subsets of a superset.

12. The system of claim 9, wherein determining that the event may occur includes: Identify multiple candidate events that are associated with both the first entity and the second entity; A filtering algorithm is used to generate a subset of candidate events associated with e-news article resources that mention both the first entity and the second entity. as well as The possibility of an event is determined based on at least one event included in a subset of candidate events.

13. The system of claim 9, wherein generating a representation corresponding to the event further comprises: Based on one or more electronic news article resources that mention both the first entity and the second entity, determine the type of event that may have occurred. The representation corresponding to an event is generated based on the type of the event.

14. The system of claim 9, wherein the representation corresponding to the event includes content describing the possible events.

15. The system of claim 9, wherein generating the second content based on additional resources comprises: The relevance between the first entity and the second entity is determined based on a subset of electronic news article resources that mention both the first entity and the second entity. Identify the attributes of specific activities included in possible events; Based on additional resources describing the correlation between the first and second entities, additional resources are selected for generating the second content; and Secondary content is generated based on the attributes of specific activities included in events related to additional resources.

16. One or more non-transitory machine-readable storage devices for storing instructions that can be executed by one or more processing devices to cause the execution of operations, said operations including: Identify the first entity based on user search queries from multiple users; In response to determining a threshold number of electronic news articles that all mention both a first entity and a second entity, wherein the magnitude of the correlation between the first entity and the second entity is at least a threshold magnitude, an event that may involve a specific activity is determined, wherein the specific activity involves both the first entity and the second entity. Generate a representation corresponding to the event. The generated representation includes: First content is generated based on one or more electronic news article resources that mention the first entity and the second entity, and Generating second content based on additional resources, wherein generating second content based on additional resources is based on determining that the additional resources are related to the attributes of specific activities included in the event. The additional resources are resources other than electronic news articles that mention the first entity and the second entity, and Among them, the attributes of a specific activity are those other than the first entity and the second entity; In response to user search queries based on multiple users, the system identifies the first entity and determines that an event relates to the first entity, and determines an interest list that includes the first entity and an event relates to the first entity: The representation is provided to the user device associated with the user account, wherein providing the representation causes the representation to be presented at the user device.

17. The one or more non-transitory machine-readable storage devices of claim 16, wherein determining that the electronic news article resources all mention both the first entity and the second entity comprises: For each electronic news article resource, generate a first resource entity score for the first entity and a second resource entity score for the second entity, and... Determine whether the scores of the corresponding first resource entity and the corresponding second resource entity satisfy the threshold.

18. The one or more non-transitory machine-readable storage devices of claim 16, wherein determining that the electronic news article resources all mention both the first entity and the second entity comprises: Identify supersets of electronic news article resources that mention both the first and second entities; For each electronic news article resource in the superset, generate the first resource entity score corresponding to the first entity and the second resource entity score corresponding to the second entity. Based on the corresponding first resource entity score and the corresponding second resource entity score, electronic news articles that mention both the first and second entities are selected from the superset. Both mention that the electronic news article resources of the first entity and the second entity are subsets of a superset.

19. The one or more non-transitory machine-readable storage devices of claim 16, wherein determining that the event may occur includes: Identify multiple candidate events that are associated with both the first entity and the second entity; A filtering algorithm is used to generate a subset of candidate events associated with e-news article resources that mention both the first entity and the second entity. as well as The possibility of an event is determined based on at least one event included in a subset of candidate events.

20. The one or more non-transitory machine-readable storage devices of claim 16, wherein generating the second content based on additional resources comprises: The relevance between the first entity and the second entity is determined based on a subset of electronic news article resources that mention both the first entity and the second entity. Determine one or more properties of possible events; Additional resources for generating the second content are selected based on additional resources describing the relationship between the first entity and the second entity. as well as The second content is generated based on the attributes of events related to the additional resources.

Images