Abstract: Among other disclosed subject matter, a computer-implemented method for determining an aspect value includes identifying a first entity associated with a first aspect that has a first value. The method includes identifying a second entity as connected to the first entity. The method includes determining a second value for a second aspect of the second entity. The method includes determining, for a third entity, a third value for the first aspect, the third value determined using at least the determined second value. Implementations can be used to evaluate truthfulness of self-reported information from members of online communities, to name one example.

Abstract: Methods, apparatuses, and systems for identifying and ranking digital resources relating to entities. A keyword included in a search query that is received, is determined to be a name of an entity. A database of names of distinct entities that uniquely identifies distinct entities that have a same name is searched. It is determined that the database associates the keyword with a first and a second entity. First and second resources, determined to be relevant to the first and the second entity, respectively, are identified from the database. In response to receiving the search query, search results directed to the first entity and separate search results directed to the second entity are provided. The search results include one or more links to the provided one or more resources.

Abstract: A server may scan an entity's or business' web page to identify information linking to one or more social networking web sites associated with the entity. The server may receive a request from a client device for information about the entity, specifically, a request to view an entity's place page. A place page may provide various types of information to users, for example, by displaying information about the particular entity such as user reviews, contact information, etc. In response to the request, the server may include the entity's social networking web site with the place page information provided to the client device. The server may also access the identified social networking web site in order to identify content such as coupons or other offers recently provided by the entity on the web site. This identified content may also be included in the place page provided to the requesting client device.

Abstract: Embodiments disclosed herein generally disclose how to provide geocoding results in response to location search queries. A geographical information system receives a search query for a map feature and identifies both web search results and geocoding results that are associated with the search query. The geographical information system determines which geocoding result to return based on the identified web search results.

Abstract: A method for determining a sentiment associated with an entity includes inputting a plurality of texts associated with the entity, labeling seed words in the plurality of texts as positive or negative, determining a score estimate for the plurality of words based on the labeling, re-enumerating paths of the plurality of words and determining a number of sentiment alternations, determining a final score for the plurality of words using only paths whose number of alternations is within a threshold, converting the final scores to corresponding z-scores for each of the plurality of words, and outputting the sentiment associated with the entity.

Abstract: Methods and apparatus, including computer program products, implementing and using techniques for providing content based on an estimated actual age. A set of related members is identified for a first member of a social networking website. Each member in the set of related members is connected to the first member in the social network website. Age information for members in the set of related members in the set of related members is examined. When a threshold number of members in the set of related members have an estimated actual age within a certain age range, an actual age of the first member is estimated based on the estimated actual age of the members in the set of related members. Content is provided to the first member based on the first member's estimated actual age. Techniques for performing a sentiment analysis based on an estimated actual age are also described.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for measuring resource geotopicality. In one aspect, a method includes receiving geotokens for a resource, where each geotoken references a geographic location. An initial geotopicality score for a geographic location is computed using token attribute values of geotokens in the resource. A set of geotopical locations are selected for the resource based on the initial geotopicality score for the geographic location. Each geotopical location can be a geographic location for which the initial geotopicality score exceeds a geotopicality threshold. A final geotopicality score representing a measure of relevance for the resource relative to the geotopical location is computed for the geotopical location. Data specifying geotopical locations for the resource and geotopicality scores for the geotopical locations are provided.

Abstract: Methods, apparatuses, and systems for identifying and ranking digital resources relating to places. A keyword included in a search query that is received, is determined to be a name of a place. A database of names of distinct places that uniquely identifies distinct places that have a same name is searched. It is determined that the database associates the keyword with a first and a second place. First and second resources, determined to be relevant to the first and the second place, respectively, are identified from the database. In response to receiving the search query, search results directed to the first place and separate search results directed to the second place are provided. The search results include one or more links to the provided one or more resources.

Abstract: A method for determining a sentiment associated with an entity includes inputting a plurality of texts associated with the entity, labeling seed words in the plurality of texts as positive or negative, determining a score estimate for the plurality of words based on the labeling, re-enumerating paths of the plurality of words and determining a number of sentiment alternations, determining a final score for the plurality of words using only paths whose number of alternations is within a threshold, converting the final scores to corresponding z-scores for each of the plurality of words, and outputting the sentiment associated with the entity.

Abstract: Methods and apparatus, including computer program products, implementing and using techniques for estimating an actual age of a member of a website. A set of related members for the member is identified. The related members are members of the same website. Age information associated with one or more related members in the set of related members is examined. When a threshold of related members in the set of related members are of an estimated actual age within a certain age range, the member's actual age is estimated to be within the age range.

Abstract: A method for determining a sentiment associated with an entity includes inputting a plurality of texts associated with the entity, labeling seed words in the plurality of texts as positive or negative, determining a score estimate for the plurality of words based on the labeling, re-enumerating paths of the plurality of words and determining a number of sentiment alternations, determining a final score for the plurality of words using only paths whose number of alternations is within a threshold, converting the final scores to corresponding z-scores for each of the plurality of words, and outputting the sentiment associated with the entity.

Abstract: A method for determining a sentiment associated with an entity includes inputting a plurality of texts associated with the entity, labeling seed words in the plurality of texts as positive or negative, determining a score estimate for the plurality of words based on the labeling, re-enumerating paths of the plurality of words and determining a number of sentiment alternations, determining a final score for the plurality of words using only paths whose number of alternations is within a threshold, converting the final scores to corresponding z-scores for each of the plurality of words, and outputting the sentiment associated with the entity.