Content recommendation based on collections of entities
First Claim
1. A method in a content recommendation computing system, the method comprising:
- presenting a first user interface control that is configured to create a new entity collection in response to received user selection of indications of named entities and suggest one or more additional entities to add to an existing collection;
presenting a second user interface control that displays information about collections created by multiple different users of the content recommendation computing system;
defining, responsive to user selections of indications of named entities via the first user interface control, a collection including multiple entities that are each referenced by one or more of multiple content items indexed by the content recommendation computing system and that are each electronically represented in the content recommendation computing system, the collection being separate from the multiple content items indexed by the content recommendation computing system;
under control of the content recommendation computing system,automatically recommending one or more additional entities to be added to the collection based on the multiple entities of the collection, by;
determining shared characteristics of the multiple entities, wherein each shared characteristic is the same for each of the multiple entities of the collection, the shared characteristics including common facets and common key terms common to the multiple entities of the collection, wherein the common facets are category values, type values, and/or characteristic values of the multiple entities and are sub-type values of a top-level entity type electronically represented by the content recommendation computing system;
determining a plurality of entities that includes at least a first entity that appears in a content item in a relationship with one of the common facets, a second entity that appears in a content item in context with one of the common key terms, and a third entity that appears in a content item in a relationship with one of the multiple entities and in context with another of the multiple entities, wherein the first entity appears in a content item in a relationship with one of the common facets when the first entity appears as a first subject or object of a subject-verb-object relation present in the content item in which the one of the common facets appears as the other of the subject or object of the first subject-verb-object relation; and
ordering the determined plurality of entities, based on entity frequency counts of frequency each entity appears in the indexed content items and on whether the determined plurality of entities share one or more of the common facets;
presenting on the first user interface control the ordered plurality of entities as suggested additions to the collection;
modifying the collection by adding one of the ordered plurality of entities to the multiple entities of the collection, the added entity selected by a user via the first user interface control;
automatically processing the modified collection to determine one or more of the multiple content items that are related to at least some of the multiple entities of the modified collection, wherein processing the modified collection includes determining whether one of the multiple entities appears in a content item in a relationship with a facet common to entities of the modified collection and in the content item in context with key terms common to the entities of the modified collection, wherein the one of the multiple entities appears in a content item in a relationship with the facet common to entities of the modified collection when the one of the multiple entities appears as a subject or object of a second subject-verb-object relation present in the content item in which the facet common to the entities of the modified collection appears as the other of the subject or object of the second subject-verb-object relation; and
recommending the determined one or more of the multiple content items.
2 Assignments
0 Petitions
Accused Products
Abstract
Techniques for content recommendation are described. Some embodiments provide a content recommendation system (“CRS”) configured to recommend content items that are related to a collection of entities. A content item may be considered related to a collection of entities based on various factors, including whether and how often the article references or otherwise covers the entities of the collection, the size of the article, other entities that are covered by the article but that are not in the collection, article recency, or article credibility. Recommending content items may also or instead include determining entities that are related to a collection. An entity can be considered related to a collection based on various factors, such as whether the entity is of the same or similar type to entities of the collection, or whether the entity appears in some article in a relationship with one or more entities of the collection.
-
Citations
34 Claims
-
1. A method in a content recommendation computing system, the method comprising:
-
presenting a first user interface control that is configured to create a new entity collection in response to received user selection of indications of named entities and suggest one or more additional entities to add to an existing collection; presenting a second user interface control that displays information about collections created by multiple different users of the content recommendation computing system; defining, responsive to user selections of indications of named entities via the first user interface control, a collection including multiple entities that are each referenced by one or more of multiple content items indexed by the content recommendation computing system and that are each electronically represented in the content recommendation computing system, the collection being separate from the multiple content items indexed by the content recommendation computing system; under control of the content recommendation computing system, automatically recommending one or more additional entities to be added to the collection based on the multiple entities of the collection, by; determining shared characteristics of the multiple entities, wherein each shared characteristic is the same for each of the multiple entities of the collection, the shared characteristics including common facets and common key terms common to the multiple entities of the collection, wherein the common facets are category values, type values, and/or characteristic values of the multiple entities and are sub-type values of a top-level entity type electronically represented by the content recommendation computing system; determining a plurality of entities that includes at least a first entity that appears in a content item in a relationship with one of the common facets, a second entity that appears in a content item in context with one of the common key terms, and a third entity that appears in a content item in a relationship with one of the multiple entities and in context with another of the multiple entities, wherein the first entity appears in a content item in a relationship with one of the common facets when the first entity appears as a first subject or object of a subject-verb-object relation present in the content item in which the one of the common facets appears as the other of the subject or object of the first subject-verb-object relation; and ordering the determined plurality of entities, based on entity frequency counts of frequency each entity appears in the indexed content items and on whether the determined plurality of entities share one or more of the common facets; presenting on the first user interface control the ordered plurality of entities as suggested additions to the collection; modifying the collection by adding one of the ordered plurality of entities to the multiple entities of the collection, the added entity selected by a user via the first user interface control; automatically processing the modified collection to determine one or more of the multiple content items that are related to at least some of the multiple entities of the modified collection, wherein processing the modified collection includes determining whether one of the multiple entities appears in a content item in a relationship with a facet common to entities of the modified collection and in the content item in context with key terms common to the entities of the modified collection, wherein the one of the multiple entities appears in a content item in a relationship with the facet common to entities of the modified collection when the one of the multiple entities appears as a subject or object of a second subject-verb-object relation present in the content item in which the facet common to the entities of the modified collection appears as the other of the subject or object of the second subject-verb-object relation; and recommending the determined one or more of the multiple content items. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A method in a content recommendation computing system, the method comprising:
-
presenting a first user interface control that is configured to create a new entity collection in response to received user selection of indications of multiple entities and suggest one or more additional entities to add to an existing collection; presenting a second user interface control that displays information about collections created by multiple different users of the content recommendation computing system; receiving, via the first user interface control, indications of multiple entities that are each referenced by one or more indexed articles and that are each electronically represented in the content recommendation computing system; under control of the content recommendation computing system, automatically recommending one or more entities based on the indicated multiple entities, by; determining shared characteristics of the multiple entities, wherein each shared characteristic is the same for each of the multiple entities, the shared characteristics including common facets and common key terms common to the multiple entities, wherein the common facets are category values, type values, and/or characteristic values of the multiple entities and are sub-type values of a top-level entity type electronically represented by the content recommendation computing system; determining a plurality of entities that includes at least a first entity that appears in an article in a relationship with one of the common facets, a second entity that appears in an article in context with one of the common key terms, and a third entity that appears in an article in a relationship with one of the multiple entities and in context with another of the multiple entities, wherein the first entity appears in an article in a relationship with one of the common facets when the first entity appears as a subject or object of a first subject-verb-object relation present in the article in which the one of the common facets appears as the other of the subject or object of the first subject-verb-object relation; and ordering the determined plurality of entities, based on entity frequency counts of frequency each entity appears in the indexed articles and on whether the determined plurality of entities share one or more of the common facets; and presenting on the first user interface control the ordered plurality of entities as suggested additions to the multiple entities; adding one of the ordered plurality of entities to the multiple entities to form modified multiple entities, the added entity selected by a user, the added entity selected by a user via the first user interface control; determining multiple articles that each include references to at least two of the modified multiple entities; determining whether one of the modified multiple entities appears in an article in a relationship with a facet common to entities of the modified multiple entities and in the article in context with key terms common to the entities of the modified multiple entities, wherein the one of the modified multiple entities appears in an article in a relationship with the facet common to entities of the modified multiple entities when the one of the modified multiple entities appears as a subject or object of a second subject-verb-object relation present in the article in which the facet common to the entities of the modified multiple entities appears as the other of the subject or object of the second subject-verb-object relation; and storing the determined multiple articles. - View Dependent Claims (10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20)
-
-
21. A method in a content recommendation computing system, the method comprising:
-
presenting a first user interface control that is configured to create a new entity collection in response to received user selection of indications of multiple named entities and suggest one or more additional entities to add to an existing collection; presenting a second user interface control that displays information about collections created by multiple different users of the content recommendation computing system; receiving from a user via the first user interface control an indication of a collection including multiple named entities that are each referenced by one or more of multiple articles indexed by the content recommendation computing system and that are each electronically represented in the content recommendation computing system, the collection being separate from the multiple articles indexed by the content recommendation computing system; and under control of the content recommendation computing system, automatically recommending one or more additional entities based on the collection to be added to the collection, by; determining shared characteristics of the multiple entities, wherein each shared characteristic is the same for each of the multiple entities of the collection, the shared characteristics including common facets and common key terms common to the multiple entities of the collection, wherein the common facets are category values, type values, and/or characteristic values of the multiple entities and are sub-type values of a top-level entity type electronically represented by the content recommendation computing system; determining a plurality of additional entities by constructing a query configured to discover relationships expressed in articles, wherein the plurality of additional entities includes at least; a first entity that appears in an article in a relationship with one of the common facets, wherein the first entity appears in the article in a relationship with one of the common facets when the first entity appears as a subject or object of a first subject-verb-object relation present in the article in which the one of the common facets appears as the other of the subject or object of the first subject-verb-object relation, a second entity that appears in an article in context with one of the common key terms, and a third entity that appears in an article in a relationship with one of the multiple entities and in the article in context with another one of the multiple entities, wherein the third entity appears in the article in a relationship with one of the multiple entities when the third entity appears as a subject or object of a second subject-verb-object relation present in the article in which the one of the multiple entities appears as the other of the subject or object of the second subject-verb-object relation; ordering the determined plurality of additional entities, based on entity frequency counts of frequency each entity appears in the indexed articles and on whether the determined additional entities share one or more of the common facets; presenting on the first user interface control the ordered plurality of entities as suggested additions to the collection; modifying the collection by adding one of the ordered plurality of entities to the collection, the added entity selected by a user via the first user interface control; automatically processing the modified collection to determine one or more of the indexed articles that are related to at least some of the multiple entities of the modified collection, wherein processing the modified collection includes determining whether one of the multiple entities appears in an article in a relationship with a facet common to entities of the modified collection and in the article in context with key terms common to the entities of the modified collection, wherein the one of the multiple entities appears in an article in a relationship with the facet common to entities of the modified collection when the one of the multiple entities appears as a subject or object of a third subject-verb-object relation present in the article in which the facet common to the entities of the modified collection appears as the other of the subject or object of the third subject-verb-object relation; storing the modified collection; and storing indications of the determined articles. - View Dependent Claims (22, 23, 24, 25, 26, 27)
-
-
28. A non-transitory computer-readable medium whose contents enable a content recommendation computing system to recommend content, by performing a method comprising:
-
presenting a first user interface control that is configured to create a new entity collection in response to received user selection of indications of named entities and suggest one or more additional entities to add to an existing collection; presenting a second user interface control that displays information about collections created by multiple different users of the content recommendation computing system; defining, responsive to user selections of indications of named entities via the first user interface control, a collection including multiple entities that are each referenced by one or more of multiple content items indexed by the content recommendation computing system and that are each electronically represented in the content recommendation computing system, the collection being separate from the multiple content items indexed by the content recommendation computing system; under control of the content recommendation computing system, automatically recommending one or more additional entities to be added to the collection based on the multiple entities of the collection, by; determining shared characteristics of the multiple entities, wherein each shared characteristic is the same for each of the multiple entities of the collection, the shared characteristics including common facets and common key terms common to the multiple entities of the collection, wherein the common facets are category values, type values, and/or characteristic values of the multiple entities and are sub-type types of a top-level entity type electronically represented by the content recommendation computing system; determining a plurality of entities that includes at least a first entity that appears in a content item in a relationship with one of the common facets, a second entity that appears in a content item in context with one of the common key terms, and a third entity that appears in a content item in a relationship with one of the multiple entities and in context with another of the multiple entities, wherein the first entity appears in a content item in a relationship with one of the common facets when the first entity appears as a subject or object of a first subject-verb-object relation present in the content item in which the one of the common facets appears as the other of the subject or object of the first subject-verb-object relation; and ordering the determined plurality of entities, based on entity frequency counts of frequency each entity appears in the indexed content items and on whether the determined plurality of entities share one or more of the common facets; presenting on the first user interface control the ordered plurality of entities as suggested additions to the collection; modifying the collection by adding one of the ordered plurality of entities to the multiple entities of the collection, the added entity selected by a user via the first user interface control; automatically processing the modified collection to determine one of the multiple content items that is related to at least some of the multiple entities of the modified collection, wherein processing the modified collection includes determining whether one of the multiple entities appears in a content item in a relationship with a facet common to entities of the modified collection and in the content item in context with key terms common to the entities of the modified collection, wherein the one of the multiple entities appears in a content item in a relationship with the facet common to entities of the modified collection when the one of the multiple entities appears as a subject or object of a second subject-verb-object relation present in the content item in which the facet common to the entities of the modified collection appears as the other of the subject or object of the second subject-verb-object relation; and recommending the determined one or more of the multiple content items. - View Dependent Claims (29, 30, 31)
-
-
32. A content recommendation computing system configured to recommend content, comprising:
-
a memory; a processing unit; a code module stored on the memory that is configured, when executed by the processing unit, to; present a user interface control that is configured to create a new entity collection in response to received user selection of indications of named entities and suggest one or more additional entities to add to an existing collection; present a second user interface control that displays information about collections created by multiple different users of the content recommendation computing system; define, responsive to user selections of indications of named entities via the first user interface control, a collection including multiple entities that are each referenced by one or more of multiple content items indexed by a content recommendation computing system and that are each electronically represented in the content recommendation computing system, the collection being separate from the multiple content items indexed by the content recommendation computing system; automatically recommend one or more additional entities to be added to the collection based on the multiple entities of the collection, by; determining shared characteristics of the multiple entities, wherein each shared characteristic is the same for each of the multiple entities of the collection, the shared characteristics including common facets and common key terms common to the multiple entities of the collection, wherein the common facets are category values, type values, and/or characteristic values of the multiple entities and are sub-type values of a top-level entity type electronically represented by the content recommendation computing system; determining a plurality of entities that includes at least a first entity that appears in a content item in a relationship with one of the common facets, a second entity that appears in a content item in context with one of the common key terms, and a third entity that appears in a content item in a relationship with one of the multiple entities and in context with another of the multiple entities, wherein the first entity appears in a content item in a relationship with one of the common facets when the first entity appears as a subject or object of a first subject-verb-object relation present in the content item in which the one of the common facets appears as the other of the subject or object of the first subject-verb-object relation; and ordering the determined plurality of entities, based on entity frequency counts of frequency each entity appears in the indexed content items and on whether the determined plurality of entities share one or more of the common facets; present on the first user interface control the ordered plurality of entities as suggested additions to the collection; modify the collection by adding one of the ordered plurality of entities to the multiple entities of the collection, the added entity selected by a user via the first user interface control; automatically process the modified collection to determine one or more content items of the multiple content items that are related to at least some of the multiple entities of the modified collection, wherein the module processes the modified collection by determining whether one of the multiple entities appears in a content item in a relationship with a facet common to entities of the modified collection and in the content item in context with key terms common to the entities of the modified collection, wherein the one of the multiple entities appears in a content item in a relationship with the facet common to entities of the modified collection when the one of the multiple entities appears as a subject or object of a second subject-verb-object relation present in the content item in which the facet common to the entities of the modified collection appears as the other of the subject or object of the second subject-verb-object relation; and recommend the determined one or more of the multiple content items. - View Dependent Claims (33, 34)
-
Specification