The processes behind the ranking of entities include: If the search result document contains information about the entities in a ranking-based order Generate a score for each entity. Scores are based on the relevance of a particular entity to a particular document. The processes behind this patent include: Receive the query. Receives information about documents related to the query. Identifies the entity associated with the document. Determine query categories based on queries, document topics, and entities. Based on the query and category, determine if you need to display the entity list in response to the query. Display search results based on the decision that you need to display the entity list in response to a query.
The search result document may contain ghost mannequin effect a list that contains information that identifies the entity. Search result documents may also contain links to documents related to the query. advertisement Continue reading below This patent is located at: Generating a ranked list of entities Inventors: Toshiaki Fujiki, Slaven Bilac, Kavi J. Goel, Shuhei Takahashi, Tomohiko Kimura Transferee : Google LLC US Patent: 10,691,702 Granted: June 23, 2020 Submitted: 2017 August 31 Overview "The device may be configured to receive queries. Receive information about the document associated with the query. Identifies the entity associated with the document. In the query, document topic, and entity. Determine the category of the query based on. Based on the query and category, determine if you need to display the entity list in response to the query, and then present the entity list in response to the query.
Presents a search result document based on the determination that there is. The search result document may contain a list that contains information that identifies the entity. " Extraction and classification of entities This patent provides an example of extracting and classifying entities from web pages and other documents. The document may contain text, images, etc. about the entity. Entities can be extracted and / or identified from a document by comparing text, images, etc. to a repository that contains information about the entity. For example, the entity may be associated with the movie "Toy Story 3". Another entity may be associated with the song "Party Rock Anthem". Another entity may be associated with the book "Galaxy Hitchhiker's Guide". Then you can classify the entities.