Microsoft Technology Licensing, LLC

Mark Blelock Atherton

Daniel Marantz

Priya Vaidyanathan

Parthasarathy Govindarajen

Garry R Wiseman

Aaron C Yuen

Nicholas Eric Craswell

Paul Baecke

Peter Richard Bailey

David Anthony Hawking

Aaron Chun-Win Yuen

Representative embodiments disclose mechanisms to complete partial queries entered by a user. Users enter a partial query. The partial query is used to search a short text index comprising the titles of documents. The search yields a list results. The top k entries of the list are selected and a language model is created from the top k entries. The language model comprises n-grams from the top k entries and an associated probability for each n-gram. A query completion generator creates query completion suggestions by matching n-grams with the partial query, removing candidate suggestions that to not comply with suggestion rules, and filtering the remaining suggestions according to a filtering criteria. The top N results are returned as suggestions to complete the query.

Corpus specific generative query completion assistant

Among other things, one or more techniques and/or systems are provided for displaying a related content view within a search interface. That is, a search interface, such as a search application, may provide search results that are relevant to a query submitted through the search interface. For example, a main search engine results view comprising one or more search results of the query may be presented within the search interface. Responsive to a semantic zoom operation (e.g., a touch gesture), the search interface may be transitioned from the main search engine results view to a related content view comprising related content that corresponds to the query and/or a (e.g., supplemental) search result for the query. In this way, the user may explore supplemental content, such as query suggestions, images, entity descriptions/profiles, videos, and/or other content, that may be related to a query submitted by the user.

Semantic zoom for related content

Architecture presents in a display related content in association with (an annotation) a social media post, as derived from post information of the social media post itself. The related content can be presented in the user interface proximate the social media post or elsewhere in the user interface. The related content can be presented as search engine queries, images, videos, and/or links to related websites. More specifically, the disclosed architecture facilitates the automated annotation of a social media post with related content derived from: post information such as a URL (uniform resource locator) referred to in the social media post, post information such as the text, image(s), and/or video content in a social media post itself, and post information such as content linked-to by a URL referred to in a social media post.

Generation of related content for social media posts

A unique system and method facilitates analyzes searches or queries that are geographic based. In particular, a user can submit a geo-tag based search for items posted on a listing service according to the item's location or geo-tag information. The geo-tag information for each item can be approximate such as the city, state, or zip code or more specific such as the street address where the item is located. The amount of geo-tag information disclosed to any other user can depend on the type of item being offered. Geo-tag information is analyzed to facilitate determining item matches. In addition, users that post items can employ query results to analyze the quality of their items such as from the buyer's perspective.

Query analysis for geographic-based listing service

A unique system and method that facilitates searching for items offered for sale or otherwise in an online classifieds environment. The system and method involve allowing a user to search for such items according to the item's location or geo-tag information. The geo-tag information for each item can be approximate such as the city, state, or zip code or more specific such as the street address where the item is located. The amount of geo-tag information disclosed to any other user can depend on the type of item being offered. One or more geo-tagged items can be returned in response to a query. The items can be returned in a list or a map format wherein the items can be pinpointed according to their locations. The map of the area around and near the item's location can provide useful information to the user to facilitate determining which item(s) to pursue.

Growing community of inventors

Lymington, United Kingdom

Mark Blelock Atherton