Yahoo! Inc.

Microsoft Technology Licensing, LLC

Akamai Technologies, Inc.

Rosie Jones

Benjamin Rey

Tina Weyand

Fernando David Diaz

Kristina Lisa Klinkner

Sarangarajan Parthasarathy

Fuchun Peng

Benoit F Dumoulin

Ahmed Hassan Awadallah

Soren Riise

Devesh Patel

Milad Shokouhi

Marco J Zagha

Eugene H Stipp

Bruce Campbell

Nick Craswell

Karthik Raghunathan

Umut Ozertem

Donald Metzler

Kevin Bartz

Omid Madani

David Richardson-Bunbury

Libin Shen

Dmitry Pechyony

Giridhar Kumaran

Desislava Petkova

Ali Ahmadi

Methods and systems are provided for providing alternative query suggestions. For example, a spoken natural language expression may be received and converted to a textual query by a speech recognition component. The spoken natural language expression may include one or more words, terms, and/or phrases. A phonetically confusable segment of the textual query may be identified by a classifier component. The classifier component may determine at least one alternative query based on identifying at least the phonetically confusable segment of the textual query. The classifier may further determine whether to suggest the at least one alternative query based on whether the at least one alternative query is sensical and/or useful. When it is determined to suggest the at least one alternative query, the at least one alternative query may be provided to and displayed on a user interface display.

Interactive reformulation of speech queries

A method to solve large scale linear SVM that is efficient in terms of computation, data storage and communication requirements. The approach works efficiently over very large datasets, and it does not require any master node to keep any examples in its memory. The algorithm assumes that the dataset is partitioned over several nodes on a cluster, and it performs 'distributed block minimization' to achieve the desired results. Using the described approach, the communication complexity of the algorithm is independent of the number of training examples.

Parallel training of a support vector machine (SVM) with distributed block minimization

A system identifies the tasks that a user is attempting to perform when submitting a series of search queries. A task may be categorized hierarchically into a mission and a goal in order to classify a series of queries and determine the user's intent. The user's mission may be an extended task and the user's goal may be a specific task. A pair of queries may be analyzed to determine whether those queries have the same mission and/or the same goal. For a series of queries, boundaries may be identified between the queries to identify where a user has changed her goal or changed her mission. The identification of goals, missions and boundaries may be automated based on an analysis of features of the queries that are compared.

Segmentation of search topics in query logs

A system and method is disclosed for determining a prediction measurement, or measure, using geo-spatial information which can be used to determine whether or not to include type of information in search results. The prediction measurement comprises a measure of the likelihood that an item of the type of information for which the prediction measure is determined will be selected, or clicked on, by a user, if the item of the type of information is included in the search result. Without limitation, one such information type is news.

System and method of geo-based prediction in search result selection

Methods and apparatus are described for identifying implicitly temporally qualified queries, i.e., queries for which a time period is implied but not explicitly stated, and for expanding such queries to include one or more temporal references.

Identifying and expanding implicitly temporally qualified queries

A familiarity level classifier comprises a stopwords engine for conducting a stopwords analysis of stopwords, e.g., introductory level stopwords and advanced level stopwords, in a document, e.g., a website; and a familiarity level classifier module for generating a document familiarity level based on the stopwords analysis. The classifier may be in an indexing module, a search engine, a user computer, or elsewhere in a computer network. The classifier may also include a reading level engine for conducting a reading level analysis of the document, and wherein the familiarity level classifier module is configured to generate the familiarity level also based on the reading level analysis. The classifier may also include a document features engine for conducting a feature analysis of the document, and wherein the familiarity level classifier module is configured to generate the document familiarity level also based on the feature analysis.

System and method for biasing search results based on topic familiarity

The present invention is directed towards systems and methods for generating one or more related queries with respect to a given query. The method may comprise receiving a query comprising one or more terms, generating a corpus of candidate reformulations and identifying one or more queries or terms from the corpus of candidate reformulations related to the received query. One or more related queries are generated on the basis of the identified one or more queries or terms from the corpus of candidate reformulations.

System and method for generating substitutable queries

A system and method are disclosed for identifying the tasks that a user is attempting to perform when submitting a series of search queries. A task may be categorized hierarchically into a mission and a goal in order to classify a series of queries and determine the user's intent. The user's mission may be an extended task and the user's goal may be a specific task. A pair of queries may be analyzed to determine whether those queries have the same mission and/or the same goal. For a series of queries, boundaries may be identified between the queries to identify where a user has changed her goal or changed her mission. The identification of goals, missions and boundaries may be automated based on an analysis of features of the queries that are compared.

A method for processing a search query having a plurality of search terms for searching for documents includes segmenting the query to identify two or more units, expanding the query by selecting one or more substitutable units for at least one unit in the query, and calculating a substitution probability for each substitutable unit. For each substitutable unit, a co-occurrence probability is calculated with each of the remaining units in the search query. An occurrence probability is then calculated for each substitutable unit, and a score is calculated based on the combination of the substitution probability, the co-occurrence probability, and occurrence probability. The documents are ranked in an order determined by the score.

Context-sensitive query expansion

A system and method for receiving a query and determining candidate alternate queries. A query may be parsed into concepts and then candidates for alternate terms may be determined for either the whole query or the determined concepts. Candidates are scored for their respective similarity to the original query. After scoring, at least one candidate is chosen and may be used as a possible alternate query choice for a user or used to serve bidded advertisements to the user.

Alternative search query processing in a term bidding system

System and method for generating substitutable queries on the basis of one or more features

The present invention relates to systems and methods for identifying one or more queries related to a given query. The method of the present invention comprises receiving a query written according to one or more writing systems of a language with multiple writing systems. A candidate set of queries written according to one or more writing systems of the language with multiple writing systems is identified. A score is calculated for the one or more queries in the candidate set indicating the similarity of the one or more queries with respect to the query received.

System and method for identifying related queries for languages with multiple writing systems

A system is disclosed for expanding a user query based on user learned popularity rankings. User queries often have no exact match, resulting in a user having to refine or abandon the search. The disclosed system obtains popularity rankings based on the content of a user query. The system expands the user query based on the popularity rankings. Thereafter, query results based on the expanded user query may be provided to the user. The system also regularly learns from user behavior and adapts the popularity rankings according to the learned information.

System and method for query expansion

System and method for determining alternate search queries

In certain implementations, a system and/or method is provided for providing search results in response to a search query. In this implementation a temporal profile of the search query is built from temporal data associated with documents retrieved in response to the search query. From features of the temporal profile it may determined whether the search query would benefit from relevance feedback from a user. If there is a determination that the search query will benefit from the relevance feedback, relevance feedback is sought from the user. Search results are provided based on the relevance feedback.

Growing community of inventors

Pasadena, CA, United States of America

Rosie Jones