Microsoft Technology Licensing, LLC

Rangan Majumder

Qifa Ke

Yu Shi

Andreas Bode

Manish Malik

Pushpraj Shukla

Krishna Chaitanya Gade

Janine Crumb

Andrew Samuel Laucius

Vishnu Challam

Javier Tordable

Ramkumar Rajendran

Non-limiting examples of the present disclosure describe a unified ranking model that may be used by a plurality of entry points to return ranked results in response to received query data. The unified ranking model is provided as a service for a plurality of entry points. A query is received from an entry point of the plurality of entry points. Results data for the query data is retrieved. A unified ranking model is executed to rank the results data. Execution of the unified ranking model manipulates feature data of the unified ranking model based on user context signals associated with the received query data and acquired result retrieval signals corresponding with the retrieved results data. Execution of the unified ranking model generates ranked result data. Ranked results data is returned to the processing device corresponding with the entry point. Other examples are also described.

Single unified ranker

Ranking domains for search engines is provided herein. To rank a domain, contributing domains associated with the domain are identified. Additionally, the maturity of each of the contributing domains is determined. A rank for the domain is then determined based at least in part on the maturity of each of the contributing domains. The domain rankings may then be used to order results for search queries.

Ranking domains using domain maturity

Systems, computer-implemented methods, and computer-readable media for modifying the rank of search result items returned by a search engine are provided. A search engine determines a plurality of search result items that satisfy a user query and the order the search result items are to be presented to a user. A rank modifier determines whether any modification should be made to the rank of each search result item identified by the search engine. The rank of search result items identified as potential spam may be demoted while the rank of search result items identified to be in the language of the search query, having a high click-through rate, or as containing adjacent search terms from the search query may be promoted. The search result items are presented according to modified ranking to the querying user.

Modifying relevance ranking of search result items

The claimed subject matter provides a system and/or a method that facilitates utilizing transition probability in static rankings associated with at least one document. An interface can receive data related to a query, wherein the query can be associated with a search from a user. A rank component can provide query results that are prioritized utilizing a transition probability based on user activity included within a user log.

UserRank: ranking linked nodes leveraging user logs

The claimed subject matter provides a system and/or a method that facilitates providing a personalized set of available content. An interface can receive a content vector associated with available content and a profile vector associated with a user preference. Additionally, a personalization component can generate a personalized set of available content by correlating the content vector and the profile vector.

Personalized search and headlines

The claimed subject matter provides systems and/or methods that enable scalable collaborative filtering. A collaborative filtering component can receive data associated with a plurality of user sessions and data associated with at least one of a user and an item. Additionally, the collaborative filtering component can generate a recommendation based on a similarity. Further, an approximation component can approximate the similarity between at least one of the item and disparate items and the user and disparate users.

Fast collaborative filtering through sketch function based approximations

Various technologies and techniques are disclosed that improve the identification of related content. An article for which to identify matching content is received or selected. The raw text of the article is analyzed to reduce the raw text to a core set of words, and the results are stored in a document feature vector array. The formatted text of the article is analyzed and vector array scores are updated based on the formatting. Anchor text words for documents that link to the article are added to the vector array. Articles linking to and from the particular article are identified and added to the vector array as appropriate. Transformations are performed, such as to adjust the vector scores based on how common or generic the words are. Vector arrays are created for other potentially related documents. The vectors are compared to determine how related they are to each other.

Growing community of inventors

Redmond, WA, United States of America

Rangan Majumder