Patents by Inventor Umut Ozertem
Umut Ozertem has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Publication number: 20200401666Abstract: Systems, methods, and non-transitory computer readable media can determine a relationship type between a first content item and a second content item based on a comment associated with at least one of the first content item and the second content item. A machine learning model can be trained based on the first content item, the second content item, and the determined relationship type. A related content item of a content item can be determined based on the machine learning model.Type: ApplicationFiled: April 17, 2018Publication date: December 24, 2020Inventors: Umut Ozertem, Ying Zhang
-
Patent number: 10423672Abstract: One embodiment identifies a set of network resources relating to a search query; determines one or more sets of query suggestions for one or more network resources from the set of network resources, respectively, wherein each one of the one or more sets of query suggestions is related to a corresponding one of the one or more network resources; and provides the one or more network resources and the one or more sets of query suggestions in response to the search query, wherein each one of the one or more sets of query suggestions is provided in association with its corresponding one of the one or more network resources.Type: GrantFiled: October 4, 2010Date of Patent: September 24, 2019Assignee: Excalibur IP, LLCInventors: Gilad Avraham Mishne, Umut Ozertem
-
Publication number: 20190266483Abstract: This application addresses techniques to de-correlate classifiers (e.g., render them neutral) to certain target groups. Classifiers can, for example, determine the intent of content (e.g., shopping, news, etc.), flag target content, etc. Sometimes, these classification categories may be incorrectly associated with certain types, groups, characteristics, etc. Exemplary embodiments retrain a classifier's model in an adversarial manner to render it no better than chance at detecting whether content originated from an entity embodying a target type, group, characteristic, etc.Type: ApplicationFiled: February 27, 2018Publication date: August 29, 2019Inventors: Umut Ozertem, Christopher Kedzie
-
Patent number: 10176219Abstract: 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.Type: GrantFiled: March 13, 2015Date of Patent: January 8, 2019Assignee: Microsoft Technology Licensing, LLCInventors: Benoit Dumoulin, Ali Ahmadi, Sarangarajan Parthasarathy, Nick Craswell, Umut Ozertem, Milad Shokouhi, Karthik Raghunathan, Rosie Jones
-
Patent number: 9947317Abstract: A new pronunciation learning system for dynamically learning new pronunciations assisted by user correction logs. The user correction logs provide a record of speech recognition events and subsequent user behavior that implicitly confirms or rejects the recognition result and/or shows the user's intended words by via subsequent input. The system analyzes the correction logs and distills them down to a set of words which lack acceptable pronunciations. Hypothetical pronunciations, constrained by spelling and other linguistic knowledge, are generated for each of the words. Offline recognition determines the hypothetical pronunciations with a good acoustical match to the audio data likely to contain the words. The matching pronunciations are aggregated and adjudicated to select new pronunciations for the words to improve general or personalized recognition models.Type: GrantFiled: February 13, 2017Date of Patent: April 17, 2018Assignee: Microsoft Technology Licensing, LLCInventors: Nicholas Kibre, Umut Ozertem, Sarangarajan Parthasarathy, Ziad Al Bawab
-
Publication number: 20170154623Abstract: A new pronunciation learning system for dynamically learning new pronunciations assisted by user correction logs. The user correction logs provide a record of speech recognition events and subsequent user behavior that implicitly confirms or rejects the recognition result and/or shows the user's intended words by via subsequent input. The system analyzes the correction logs and distills them down to a set of words which lack acceptable pronunciations. Hypothetical pronunciations, constrained by spelling and other linguistic knowledge, are generated for each of the words. Offline recognition determines the hypothetical pronunciations with a good acoustical match to the audio data likely to contain the words. The matching pronunciations are aggregated and adjudicated to select new pronunciations for the words to improve general or personalized recognition models.Type: ApplicationFiled: February 13, 2017Publication date: June 1, 2017Applicant: Microsoft Technology Licensing, LLC.Inventors: Nicholas Kibre, Umut Ozertem, Sarangarajan Parthasarathy, Ziad Al Bawab
-
Patent number: 9589562Abstract: A new pronunciation learning system for dynamically learning new pronunciations assisted by user correction logs. The user correction logs provide a record of speech recognition events and subsequent user behavior that implicitly confirms or rejects the recognition result and/or shows the user's intended words by via subsequent input. The system analyzes the correction logs and distills them down to a set of words which lack acceptable pronunciations. Hypothetical pronunciations, constrained by spelling and other linguistic knowledge, are generated for each of the words. Offline recognition determines the hypothetical pronunciations with a good acoustical match to the audio data likely to contain the words. The matching pronunciations are aggregated and adjudicated to select new pronunciations for the words to improve general or personalized recognition models.Type: GrantFiled: February 21, 2014Date of Patent: March 7, 2017Assignee: Microsoft Technology Licensing, LLCInventors: Nicholas Kibre, Umut Ozertem, Sarangarajan Parthasarathy, Ziad Al Bawab
-
Patent number: 9460081Abstract: Examples of the present disclosure describe generation of a multi-arc confusion network to improve, for example, an ability to return alternatives to output generated. A confusion network comprising token representations of lexicalized hypotheses and normalized hypotheses is generated. Each arc of the confusion network represents a token of a lexicalized hypothesis or a normalized hypothesis. The confusion network is transformed into a multi-arc confusion network, wherein the transforming comprising realigning at least one token of the confusion network to span multiple arcs of the confusion network. Other examples are also described.Type: GrantFiled: June 2, 2016Date of Patent: October 4, 2016Assignee: Microsoft Technology Licensing, LLCInventors: Michael Levit, Umut Ozertem, Sarangarajan Parthasarathy, Padma Varadharajan, Karthik Raghunathan, Issac Alphonso
-
Publication number: 20160275071Abstract: Examples of the present disclosure describe generation of a multi-arc confusion network to improve, for example, an ability to return alternatives to output generated. A confusion network comprising token representations of lexicalized hypotheses and normalized hypotheses is generated. Each arc of the confusion network represents a token of a lexicalized hypothesis or a normalized hypothesis. The confusion network is transformed into a multi-arc confusion network, wherein the transforming comprising realigning at least one token of the confusion network to span multiple arcs of the confusion network. Other examples are also described.Type: ApplicationFiled: June 2, 2016Publication date: September 22, 2016Applicant: Microsoft Technology Licensing, LLCInventors: Michael Levit, Umut Ozertem, Sarangarajan Parthasarathy, Padma Varadharajan, Karthik Raghunathan, Issac Alphonso
-
Publication number: 20160267128Abstract: 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.Type: ApplicationFiled: March 13, 2015Publication date: September 15, 2016Applicant: Microsoft Technology Licensing , LLCInventors: Benoit Dumoulin, Ali Ahmadi, Sarangarajan Parthasarathy, Nick Craswell, Umut Ozertem, Milad Shokouhi, Karthik Raghunathan, Rosie Jones
-
Publication number: 20160217125Abstract: Examples of the present disclosure describe generation of a multi-arc confusion network to improve, for example, an ability to return alternatives to output generated. A confusion network comprising token representations of lexicalized hypotheses and normalized hypotheses is generated. Each arc of the confusion network represents a token of a lexicalized hypothesis or a normalized hypothesis. The confusion network is transformed into a multi-arc confusion network, wherein the transforming comprising realigning at least one token of the confusion network to span multiple arcs of the confusion network. Other examples are also described.Type: ApplicationFiled: January 27, 2015Publication date: July 28, 2016Applicant: Microsoft Technology Licensing, LLCInventors: Michael Levit, Umut Ozertem, Sarangarajan Parthasarathy, Padma Varadharajan, Karthik Raghunathan, Issac Alphonso
-
Patent number: 9384188Abstract: Examples of the present disclosure describe generation of a multi-arc confusion network to improve, for example, an ability to return alternatives to output generated. A confusion network comprising token representations of lexicalized hypotheses and normalized hypotheses is generated. Each arc of the confusion network represents a token of a lexicalized hypothesis or a normalized hypothesis. The confusion network is transformed into a multi-arc confusion network, wherein the transforming comprising realigning at least one token of the confusion network to span multiple arcs of the confusion network. Other examples are also described.Type: GrantFiled: January 27, 2015Date of Patent: July 5, 2016Assignee: Microsoft Technology Licensing, LLCInventors: Michael Levit, Umut Ozertem, Sarangarajan Parthasarathy, Padma Varadharajan, Karthik Raghunathan, Issac Alphonso
-
Publication number: 20150243278Abstract: A new pronunciation learning system for dynamically learning new pronunciations assisted by user correction logs. The user correction logs provide a record of speech recognition events and subsequent user behavior that implicitly confirms or rejects the recognition result and/or shows the user's intended words by via subsequent input. The system analyzes the correction logs and distills them down to a set of words which lack acceptable pronunciations. Hypothetical pronunciations, constrained by spelling and other linguistic knowledge, are generated for each of the words. Offline recognition determines the hypothetical pronunciations with a good acoustical match to the audio data likely to contain the words. The matching pronunciations are aggregated and adjudicated to select new pronunciations for the words to improve general or personalized recognition models.Type: ApplicationFiled: February 21, 2014Publication date: August 27, 2015Applicant: MICROSOFT CORPORATIONInventors: Nicholas Kibre, Umut Ozertem, Sarangarajan Parthasarathy, Ziad Al Bawab
-
Patent number: 8930338Abstract: Disclosed is a system and method for providing search suggestions to a user based on the user's previously entered search queries. A computing device stores a global set of search suggestions. The computing device receives over a network from a user computer operated by a user one or more alphanumeric characters forming a portion of a search query. The computing device determines a search suggestion to the portion of the search query from the global set of search suggestions based on a search history of the user, the search history of the user comprising a plurality of search queries entered by the user within a predetermined period of time. The computing device transmits to the user computer the search suggestion for display by the user computer.Type: GrantFiled: May 17, 2011Date of Patent: January 6, 2015Assignee: Yahoo! Inc.Inventors: Omer Emre Velipasaoglu, Umut Ozertem, Alpa Jain
-
Publication number: 20130132433Abstract: A method and system for categorizing web-search queries in semantically coherent topics. The method includes receiving plurality of web-search queries from one or more users and storing the plurality of web-search queries in a query log. The method further includes processing the plurality of web-search queries for topic generation by generating plurality of missions from the query log and merging together one or more missions belonging to a similar topic. Further, the method includes determining topical user profile of a user by matching each mission of the user with one or more relevant topics, and detecting user activity of the user from random user activity. Moreover, the method includes naming one or more semantically coherent topics using a set of common concept terms extracted from the plurality of web-search queries. The system includes one or more electronic devices, a communication interface, a memory, and a processor.Type: ApplicationFiled: November 22, 2011Publication date: May 23, 2013Applicant: Yahoo! Inc.Inventors: Umut Ozertem, Debora Donato, Luca Aiello
-
Publication number: 20120296743Abstract: Method, system, and programs for providing personalized suggest-as-you-type suggestions in response to a user search query wherein the personalized query suggestions are based on the user's past interactions with the system. The system is able to identify frequent queries issued by the user that result in the user clicking on the same universal resource locator.Type: ApplicationFiled: May 19, 2011Publication date: November 22, 2012Applicant: YAHOO! INC.Inventors: Omer Emre Velipasaoglu, Umut Ozertem
-
Publication number: 20120296927Abstract: Disclosed is a system and method for providing search suggestions to a user based on the user's previously entered search queries. A computing device stores a global set of search suggestions. The computing device receives over a network from a user computer operated by a user one or more alphanumeric characters forming a portion of a search query. The computing device determines a search suggestion to the portion of the search query from the global set of search suggestions based on a search history of the user, the search history of the user comprising a plurality of search queries entered by the user within a predetermined period of time. The computing device transmits to the user computer the search suggestion for display by the user computer.Type: ApplicationFiled: May 17, 2011Publication date: November 22, 2012Applicant: Yahoo! Inc.Inventors: Omer Emre Velipasaoglu, Umut Ozertem, Alpa Jain
-
Publication number: 20120191745Abstract: Data-mining software receives a user query as an input and segments the user query into a number of units. The data-mining software then drops terms from a unit using a Conditional Random Field (CRF) model that combines a number of features. At least one of the features is derived from query logs and at least one of the features is derived from web documents. The data-mining software then generates one or more candidate queries by adding terms to the unit. The added terms result from a hybrid method that utilizes query sessions and a web corpus. The data-mining software also scores each candidate query on well-formedness of the candidate query, utility, and relevance to the user query. Then the data-mining software stores the scored candidate queries in a database for subsequent display in a graphical user interface for a search engine.Type: ApplicationFiled: January 24, 2011Publication date: July 26, 2012Applicant: Yahoo!, Inc.Inventors: Emre Velipasaoglu, Alpa Jain, Umut Ozertem
-
Publication number: 20120084297Abstract: One embodiment identifies a set of network resources relating to a search query; determines one or more sets of query suggestions for one or more network resources from the set of network resources, respectively, wherein each one of the one or more sets of query suggestions is related to a corresponding one of the one or more network resources; and provides the one or more network resources and the one or more sets of query suggestions in response to the search query, wherein each one of the one or more sets of query suggestions is provided in association with its corresponding one of the one or more network resources.Type: ApplicationFiled: October 4, 2010Publication date: April 5, 2012Applicant: YAHOO! INC.Inventors: Gilad Avraham Mishne, Umut Ozertem