Based On Record Similarity And Relevance Patents (Class 707/749)
-
Patent number: 8838610Abstract: A listing tune-up system is provided. An example listing tune-up system may include a detecting module to detect a listing for an item. The listing may comprise various information including a title, and a category of the item for transaction. The example listing tune-up system may include a determining module to determine a recommended category set for the item using categories of existing listings that match one or more keywords in the title. The example listing tune-up system may include a verifying module to verify whether the category of the item complies with the recommended category set. The example listing tune-up system may further include a generating module to generate a tune-up report for the listing upon completion of the verification. The tune-up report may comprise the recommended category set and/or other editing suggestions.Type: GrantFiled: October 30, 2009Date of Patent: September 16, 2014Assignee: eBay Inc.Inventor: Jason Adam Swartz
-
Patent number: 8838453Abstract: A user input is received by a computing device. An interactive input module determines whether the first user input is a first character of a script for a supported language. If the first user input is a first character, the first character is stored in an input buffer. A plurality of words in the supported language that match a contents of the input buffer are identified, and a subset of the plurality of words are displayed to the user based on a frequency value associated with each of the plurality of words.Type: GrantFiled: August 31, 2010Date of Patent: September 16, 2014Assignee: Red Hat, Inc.Inventor: Pravin Satpute
-
Patent number: 8838616Abstract: A server device of the present invention includes a control unit collecting texts stored in a storage unit in response to an instruction from the outside or when a predetermined time is reached, extracting words from the collected texts, determining, as a general word, a word which appears at a frequency higher than a first predefined value for a first predetermined period, and which appears at a frequency that varies within a second predefined value range for every second predetermined period that is shorter than the first predetermined period, and creating a general word list which enumerates the general words.Type: GrantFiled: August 18, 2009Date of Patent: September 16, 2014Assignee: NEC Biglobe, Ltd.Inventor: Norikazu Matsumura
-
Systems and methods for classifying electronic information using advanced active learning techniques
Patent number: 8838606Abstract: Systems and methods for classifying electronic information or documents into a number of classes and subclasses are provided through an active learning algorithm. In certain embodiments, seed sets may be eliminated by merging relevance feedback and machine learning phases. Such document classification systems are easily scalable for large document collections, require less manpower and can be employed on a single computer, thus requiring fewer resources. Furthermore, the classification systems and methods described can be used for any pattern recognition or classification effort in a wide variety of fields, including electronic discovery in legal proceedings.Type: GrantFiled: June 18, 2013Date of Patent: September 16, 2014Inventors: Gordon Villy Cormack, Maura Robin Grossman -
Publication number: 20140258310Abstract: A method of operation of a navigation system includes: generating a point of interest record based on normalizing an entry record; calculating a similarity score based on comparing a plurality of the POI record to each other; and generating an affinity group with a control unit based on the similarity score meeting or exceeding an edge weight threshold for displaying on a device.Type: ApplicationFiled: March 5, 2013Publication date: September 11, 2014Applicant: TELENAV, INC.Inventors: Jaffe Wong, Aliasgar Mumtaz Husain, Sunny Dinmohamed Panjwani
-
Publication number: 20140258312Abstract: Techniques are disclosed for determining reasons underlying insights gleaned from multi-dimensional data. In one embodiment, a contingency table is accessed that represents multiple dimensions of the data, in order to identify one or more insights. One or more dimensions, other than the represented dimensions, are evaluated to identify one or more reasons underlying a first insight of the one or more insights, and the one or more reasons are output.Type: ApplicationFiled: March 25, 2014Publication date: September 11, 2014Applicant: International Business Machines CorporationInventors: Felix HAMBORG, Alexander LANG, Christoph LINGENFELDER
-
Publication number: 20140258311Abstract: Techniques are disclosed for determining reasons underlying insights gleaned from multi-dimensional data. In one embodiment, a contingency table is accessed that represents multiple dimensions of the data, in order to identify one or more insights. One or more dimensions, other than the represented dimensions, are evaluated to identify one or more reasons underlying a first insight of the one or more insights, and the one or more reasons are output.Type: ApplicationFiled: March 7, 2013Publication date: September 11, 2014Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: Felix Hamborg, Alexander Lang, Christoph Lingenfelder
-
Publication number: 20140258305Abstract: The disclosed embodiments include methods and systems for providing contextual trust scores. The disclosed embodiments include, for example, a system for providing a contextual trust score including a memory storing software instructions and one or more processors configured to execute the software instructions. In one aspect, the one or more processors may be configured to perform operations including receiving a user scenario associated with a user. The operations may also include selecting one or more key variables from one or more data sources based on the user scenario, and may measure the key variables across one or more contextual dimensions. The operations may further include comparing the results of the measuring, and generating a contextual trust score associated with a user. The operations may also include continuously monitoring the data sources, updating the generated trust score, and providing the update trust score to the user or a third party.Type: ApplicationFiled: March 5, 2014Publication date: September 11, 2014Applicant: TREMUS, INC. D/B/A TRUSTFACTORS, INC.Inventors: Kalpesh Gopaldas KAPADIA, Michael Rourk HALLINAN
-
Patent number: 8832119Abstract: An adaptive, interactive visual workspace for viewing groups of files based on their relationships. Relationships of files are visualized using iterative refinement of categories through a direct-manipulation graph-based layout. The visual workspace starts with a fully connected graph linking thumbnail images of related files that is then partitioned into neighborhoods in response to a user creating file stacks corresponding to different categories. Normalized spring lengths improve the overall quality of the layout. Different modes for membership in neighborhoods avoid confusing motion of files and help a user to manually organize the workspace. Additionally, retrieved files can be added without having to significantly move the previous files. Different visualization techniques indicate which files are related to each other. Different zoom rates are used for file location, and surrogate sizes allow users to increase the separation between files while still increasing the surrogate sizes.Type: GrantFiled: June 12, 2008Date of Patent: September 9, 2014Assignee: Fuji Xerox Co., Ltd.Inventors: Andreas Girgensohn, Frank Shipman, III, Lynn Wilcox
-
Patent number: 8832102Abstract: Exemplary methods and apparatuses are provided which may be implemented using one or more computing devices to allow for super clustering of clusters of electronic documents based, at least in part, on structural and static content features.Type: GrantFiled: January 12, 2010Date of Patent: September 9, 2014Assignee: Yahoo! Inc.Inventors: Rupesh R. Mehta, Srinivasan H. Sengamedu, Rajeev R. Rastogi
-
Patent number: 8832126Abstract: Given a set of documents relevant to a litigation hold and a seed set of custodians, a second set of custodians can be generated and suggested to a user. After receiving a seed set of keywords and/or custodians, documents are identified based on their relevance. Relevant documents are searched for custodian names, and appropriate custodian names are presented to a user. Additionally, based on a first set of custodians, a suggested set of custodians can be provided to a user based on relationships between the sets of custodians.Type: GrantFiled: August 29, 2011Date of Patent: September 9, 2014Assignee: Google Inc.Inventors: Shailesh Kumar, Mahesh Chhaparia, Gaurav Agarwal
-
Patent number: 8832120Abstract: A computer-based method of determining a weirdness score for variables within a data set is provided. The method includes receiving a selection of a first variable, wherein the first variable is defined by a measure, a time period, and a plurality of entities, calculating a plurality of parameters for the first variable, wherein each of the plurality of parameters is calculated based at least in part on a deviation of a measured value from a predicted value, calculating a rank for each of the plurality of parameters for the first variable, wherein the rank of each parameter is calculated relative to corresponding parameters calculated for all other variables in the data set having the same measure and time period as the first variable, and calculating a weirdness score for the first variable based at least in part on the calculated rank of each of the plurality of parameters.Type: GrantFiled: January 30, 2012Date of Patent: September 9, 2014Assignee: MasterCard International IncorporatedInventor: William C. Jones
-
Patent number: 8832108Abstract: Classifying documents that have different scales is described. Instances are counted for each character size in documents. Character sizes for the first document and the second document are selected based on the instance count for each character size. Scales are calculated based on ratios of each first character size relative to each second character size. Scale products are calculated based on each instance count for each character size range for the first character sizes multiplied by each instance count for each corresponding character size range for the second character sizes. The corresponding character size range is based on a corresponding scale. Scale scores are calculated based on summing each of the scale products for each scale. A scale is selected based a highest scale score. The second document may be classified with the first document based on a comparison of first document location information and second document location information.Type: GrantFiled: April 18, 2013Date of Patent: September 9, 2014Assignee: EMC CorporationInventor: Steven Sampson
-
Publication number: 20140250087Abstract: A computer-implemented system and method for identifying relevant documents for display are provided. Themes for a set of documents are generated. The documents are clustered based on the themes. A matrix including an inner product of document frequency occurrences and cluster concept weightings for each theme is generated for the documents. From the matrix, documents most relevant to a particular theme are identified, and the relevant documents are displayed.Type: ApplicationFiled: May 12, 2014Publication date: September 4, 2014Applicant: FTI TECHNOLOGY LLCInventors: Dan Gallivan, Kenji Kawai
-
Patent number: 8825673Abstract: A computer product including a data structure for organizing of a plurality of documents, and capable of being utilized by a processor for manipulating data of the data structure and capable of displaying selected data on a display unit. The data structure includes a plurality of directionally interlinked nodes, each node being associated with one or more documents having a header and body text. All the documents are associated with a given node and have identical normalized body text. All documents that have identical normalized body text are associated with the same node. One or more of the nodes is associated with more than one document. For any node that is a descendent of another node, the normalized body text of each document associated with the node is inclusive of the normalized body text of a document that is associated with the other node.Type: GrantFiled: July 2, 2008Date of Patent: September 2, 2014Assignee: Equivio Ltd.Inventors: Yiftach Ravid, Amir Milo
-
Patent number: 8825645Abstract: A ranking component ranks documents, such as web pages or web sites, to obtain a ranking score that defines a quality judgment of the document. The ranking score of a particular document is based on the ranking score of the documents which link to it and based on affiliation among the documents.Type: GrantFiled: January 6, 2012Date of Patent: September 2, 2014Assignee: Google Inc.Inventors: Krishna Bharat, Amit Singhal, Paul Haahr
-
Patent number: 8825574Abstract: Peer-to-peer media item recommendations are provided by presenting a selected list of media items to a first user to consider for recommendation to a friend, receiving an indication from the first user that one or more of the presented media items are to be recommended to the friend, and presenting the one or more indicated media items to the friend. The presented list comprises media items the friend hasn't rated, and is selected using a media recommendation engine.Type: GrantFiled: September 28, 2010Date of Patent: September 2, 2014Inventors: Robert Bodor, Aaron Weber
-
Publication number: 20140244664Abstract: A method for determining similarity information for users of a network is disclosed. The method comprises constructing a graph from user interaction data, wherein the graph links users of the network according to their interaction, and weighting links between users according to the strength of interaction between the users. The method further comprises, for each of a plurality of users, identifying from among the other users with whom the user interacts: a maximum similarity user to whom the user is connected via a maximum weight link, and a minimum similarity user to whom the user is connected via a minimum weight link. Finally the method comprises assembling a plurality of relative similarity triplets, each triplet comprising a user and that user's identified maximum and minimum similarity users.Type: ApplicationFiled: February 25, 2013Publication date: August 28, 2014Applicant: Telefonaktiebolaget L M Ericsson (publ)Inventor: Vikas VERMA
-
Publication number: 20140244665Abstract: In one embodiment, location-based names are selected for a digital object based on a user's familiarity with the geographic location where the digital object was generated. For example, a memory stores multiple names indicative of the same geographic location. The multiple names may include a street address, a city, and a state, where each defines a different level of location specificity. The device that generates the digital object also collects familiarity data for the geographic location and calculates a familiarity score for the digital object based on the familiarity data for the user or the device. If the familiarity score is above a threshold, the user of the device is familiar with the geographic location, and a specific name is associated with the digital object. If the familiarity score is below the threshold, the user of the device is not familiar with the geographic location, and a broader name is associated with the digital object.Type: ApplicationFiled: February 27, 2013Publication date: August 28, 2014Applicant: NAVTEQ B.V.Inventors: Stephen Andrew Hartford, Albert William Biggerstaff
-
Patent number: 8818978Abstract: In some example embodiments, a system and method are illustrated to associate an item listing with one or more corresponding images. The system and method include receiving an item listing for an item from a user device. The system and method include generating a similarity score for a respective existing image associated with one or more existing item listings. The generating may be done by comparing the item listing with the existing item listings received from the user device. The similarity score may indicate a degree of similarity between the item listing and the existing item listings associated with the respective existing image. The system and method include proposing a specified number of existing images with the highest similarity scores to the user device. The system and method further include associating the item listing with one or more of the specified number of existing images accepted by the user device.Type: GrantFiled: August 15, 2008Date of Patent: August 26, 2014Assignee: eBay Inc.Inventors: Xiaoyuan Wu, Alvaro Bolivar, Qiang Wang
-
Patent number: 8818337Abstract: A method of organizing mobile content in a network environment is provided that includes providing pieces of mobile content on a database, selecting one of the pieces of mobile content, receiving a descriptor to be associated with the selected piece of mobile content from user activity, and associating the descriptor to the selected piece of mobile content.Type: GrantFiled: December 28, 2006Date of Patent: August 26, 2014Assignee: FunMobility, Inc.Inventors: Srivathsan Narasimhan, Eric F. Allen, Skot Leach, Hudson George, Lincoln Lydick, Yu-Jen Dennis Chen, Adam Lavine, Silvy Mathews
-
Patent number: 8818979Abstract: A document retrieving apparatus can retrieve a target document and output the retrieved target documents according to ranking when a retrieval keyword or retrieval expression is input. However, it requires a skilful technique to narrow a retrieval range since an appropriate retrieval keyword or retrieval expression needs to be created. A document retrieving apparatus of the present invention reads out and compiles a document list included in a designated area when a user designates an area of a document to be read on a two-dimensional map. When the user designates an area of a document to be read on the two-dimensional map, the document retrieving apparatus of the present invention combines query vectors of a plurality of documents included in a designated area and extracts documents based on a combined query vector.Type: GrantFiled: May 15, 2008Date of Patent: August 26, 2014Assignee: Valuenex Consulting Inc.Inventors: Tatsuo Nakamura, Yoshio Takaeda
-
Patent number: 8819047Abstract: The described implementations relate to processing of electronic data. One implementation is manifested as a technique that can include receiving an input statement that includes a plurality of terms. The technique can also include providing, in response to the input statement, ranked supporting documents that support the input statement or ranked contradicting results that contradict the input statement.Type: GrantFiled: April 4, 2012Date of Patent: August 26, 2014Assignee: Microsoft CorporationInventors: Silviu-Petru Cucerzan, Chee Wee Leong
-
Publication number: 20140236969Abstract: An apparatus and method for recommending contents based on a metadata graph are provided. The apparatus includes a contents metadata acquirer configured to acquire metadata of contents, a contents-related data acquirer configured to acquire contents-related data from the Web, a metadata graph generator configured to analyze acquired contents metadata and contents-related data to generate a metadata graph, a metadata graph DB configured to store a metadata graph generated by the metadata graph generator, and a contents recommender configured to search for contents that are similar to specific contents viewed by a user through a contents output terminal using the metadata graph stored in the metadata graph DB, and recommend the searched contents to the user through the contents output terminal.Type: ApplicationFiled: June 28, 2013Publication date: August 21, 2014Applicant: Electronics and Telecommunications Research InstituteInventors: Goong-Hyun NAM, Kang-Yong LEE, Cho-Rong YU, Kee-Seong CHO
-
Patent number: 8812494Abstract: Systems and methods are provided for selecting contexts for new invitational content and invitational content for new contexts. In particular, a performance history of delivered invitational content in known contexts is combined with similarity measures for the delivered invitational content, with respect to a new invitational content, to generate a list of potential contexts for the new invitational content. Similarly, a performance history of in known contexts with delivered invitational content can combined with similarity measures for known contexts, with respect to a new context, to generate a list of potential content for the new context. Further, a combination of these methods can be used to pair new invitational content with new contexts.Type: GrantFiled: January 4, 2013Date of Patent: August 19, 2014Assignee: Apple Inc.Inventors: Eswar Priyadershan, Dan Marius Grigorovici, Ravikiran Chittari, Prasenjit Mukherjee, Omar Abdala
-
Publication number: 20140229494Abstract: Techniques for sharing item images based on a similarity score are described. For example, a machine receives an item listing for an item from a user device. The machine generates a similarity score for an existing image corresponding to one or more existing item listings by comparing the item listing received from the user device with the one or more existing item listings. The similarity score may indicate a degree of similarity between the item listing received from the user device and the one or more existing item listings corresponding to the existing image. The machine selects one or more existing images corresponding to the one or more existing item listings based on the one or more existing images having corresponding similarity scores that exceed a threshold value. The machine transmits a communication to the user device in response to the receiving of the item listing from the user device. The communication may include the one or more existing images.Type: ApplicationFiled: April 16, 2014Publication date: August 14, 2014Applicant: eBay Inc.Inventors: Xiaoyuan Wu, Alvaro Bolivar, Qiang Wang
-
Patent number: 8805854Abstract: In an embodiment, a method is provided for determining a mood profile of media data. In this method, mood is determined across multiple elements of mood for the media data to create a mood profile associated with the media data. In some embodiments, the mood profile is then used to determine congruencies between one or more pieces of media data.Type: GrantFiled: June 23, 2009Date of Patent: August 12, 2014Assignee: Gracenote, Inc.Inventors: Ching-Wei Chen, Kyogu Lee, Peter C. DiMaria, Markus K. Cremer
-
Patent number: 8805855Abstract: For efficiently storing and retrieving data and metadata in phases, in a first phase, metadata tokens, which are assigned to most active metadata-emitting entities, are used for storing the data and the metadata together in a single input/output operation while piggybacking the metadata of least active metadata-emitting entities onto one of the most active metadata-emitting entities having one of the metadata tokens. In a second phase, the metadata is re-written to a metadata delta journal for reclaiming the metadata tokens. In a third phase, the metadata journal is applied to a metadata structure containing the metadata of the storage system, the metadata delta journal is then cleared after successfully updating the main metadata structure with the metadata of the metadata journal. The metadata journal is swapped with an empty metadata journal for concurrently adding metadata while retaining the metadata journal until applying the metadata delta journal to the metadata structure.Type: GrantFiled: August 17, 2012Date of Patent: August 12, 2014Assignee: International Business Machines CorporationInventors: Ben Sasson, Ori Shalev
-
Patent number: 8805843Abstract: A method and analytics tools for information mining incorporating domain specific knowledge and conceptual structures are disclosed, the method including: providing a first set of documents related to a first topic of interest; using a first taxonomy to categorize the first set of documents into a set of categories; providing a second set of documents related to a second topic of interest; categorizing the second set of documents according to the set of categories of the first set of documents; using an element of domain knowledge to re-categorize the first set of documents; and examining a category to identify a document of interest.Type: GrantFiled: June 3, 2008Date of Patent: August 12, 2014Assignee: International Business Machines CorporationInventors: Ying Chen, Jeffrey Thomas Kreulen, James J. Rhodes, William Scott Spangler
-
Patent number: 8799455Abstract: A system and a method for the management of addressable network resources (e.g., Web pages, electronic documents accessed via a Uniform Resource Identifier, etc.) within a network are provided. Specifically, embodiments of addressable network resource selection systems and methods are disclosed for managing the analysis and identification of optimal links between addressable network resources. A resource selection component enables pruning of undesirable pages on the basis of various usage metrics, including user depart rate. The resource selection component may then identify optimal linking strategies by determining one or more linking scores for remaining pages based on any of a number of different page usage metrics. In other embodiments, addressable network resource selection systems and methods may weight linking scores to emphasize desired outcomes, such as encouraging exploration of a network-based site or encouraging high aggregate sales figures.Type: GrantFiled: March 18, 2011Date of Patent: August 5, 2014Assignee: Amazon Technologies, Inc.Inventors: Jinesh U. Vora, Adam Brent Johnson
-
Publication number: 20140214861Abstract: A social data aggregator generates entries of action data describing actions taken by users of a social networking system. The entries are archived in a number of remote data hosts, with a portion of the entries stored in an action cache to expedite retrieval. The action cache is modified to include relevant entries of action data by selecting the most relevant entries from proxy data aggregators, which identify sets of entries from remote data hosts coupled to the proxy data aggregators based on the relevance of the entries in the remote data hosts. The relevance of an entry is based on an associated engagement score indicating a likelihood of a user requesting content interacting with a notification based on an entry.Type: ApplicationFiled: January 31, 2013Publication date: July 31, 2014Applicant: Facebook, Inc.Inventors: Li Zheng, Ankush Singla
-
Publication number: 20140214737Abstract: The invention provides a display interface in a social networking system that enables the presentation of information related to a user in a timeline or map view. The system accesses information about a user of a social networking system, including both data about the user and social network activities related to the user. The system then selects one or more of these pieces of data and/or activities from a certain time period and gathers them into timeline units based on their relatedness and their relevance to users. These timeline units are ranked by relevance to the user, and are used to generate a timeline or map view for the user containing visual representations of the timeline units organized by location or time. The timeline or map view is then provided to other users of the social networking system that wish to view information about the user.Type: ApplicationFiled: March 26, 2014Publication date: July 31, 2014Applicant: Facebook, Inc.Inventors: Serkan Piantino, Ryan Case, Stanislav Funiak, Daniel Klatzko Gibson, Jeff Huang, Ryan David Mack, Paul M. McDonald, Arun Vijayvergiya, Joshua Wiseman, Zishuang Yang, Steven Young, Raylene Kay Yung, Mark E. Zuckerberg
-
Patent number: 8793260Abstract: Systems and methods for responding to search queries with search results and related pivoted search queries is presented. Responsive to a search query, one or more entities related to the search query are identified as pivot entities. For each pivot entity a set of pivoted search queries is identified and included among the search results responsive to the search query.Type: GrantFiled: April 5, 2012Date of Patent: July 29, 2014Assignee: Microsoft CorporationInventors: Salem Elie Haykal, Daniel Thomas Herron
-
Patent number: 8793259Abstract: In an exemplary aspect, the present invention includes a control unit that when a keyword for search is entered, collects texts containing that keyword from texts stored in a storage unit, extracts a noun of collected first texts, determines a noun partially matching with the keyword as a first word, extracts a second text containing that first word among the first texts, extracts a word from the second text, the word being one of a noun, a verb, and an adjective, counts the number of times an extracted word is used, determines a word whose number of times of use is placed in predefined highest ranks as a second word, the second word being a related word to the first word, and outputs the first word and the second word.Type: GrantFiled: August 18, 2009Date of Patent: July 29, 2014Assignee: NEC Biglobe, Ltd.Inventor: Norikazu Matsumura
-
Publication number: 20140207800Abstract: Various embodiments select features from a feature space. In one embodiment a candidate feature set of k? features is selected from at least one set of features based on maximum relevancy and minimum redundancy (MRMR) criteria. A target feature set of k features is identified from the candidate feature set, where k?>k. Each a plurality of features in the target feature set is iteratively updated with each of a plurality of k??k features from the candidate feature set. The feature from the plurality of k??k features is maintained in the target feature set, for at least one iterative update, based on a current MRMR score of the target feature set satisfying a threshold. The target feature set is stored as a top-k feature set of the at least one set of features after a given number of iterative updates.Type: ApplicationFiled: September 18, 2013Publication date: July 24, 2014Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: David HAWS, Dan HE, Laxmi P. PARIDA
-
Publication number: 20140207802Abstract: Highly relevant search results for unstructured queries are efficiently identified from normalized data. A server graphs relationships between each of the data objects. For each search term, the server identifies a candidate set of data objects mapped to the term. The server calculates priority scores for the candidate data objects based at least in part on one or more of: a link analysis of the graph; or an analysis of metadata describing structural constraints upon the candidate data objects. Based on the graph, the server identifies one or more search result subgraphs, each comprising at least one data object from each of the candidate sets. The server looks for subgraphs in an order that is based on the priority scores for the candidate data objects. The server may terminate its search early, in part because prioritizing the candidate data objects increases the likelihood of receiving relevant search results first.Type: ApplicationFiled: March 21, 2014Publication date: July 24, 2014Applicant: Oracle International CorporationInventors: Nikhil Raghavan, Ravi Murthy, Aman Naimat
-
Publication number: 20140207746Abstract: When a user-submitted query is received, a set of candidate queries is identified. For each of the candidate queries, features are extracted that, for each candidate query, reflect a measure of effectiveness of the candidate query. The candidate queries are rank ordered based on the measure of effectiveness, and one or more of the top-ranked candidate queries are presented as suggested alternatives to the user-submitted query.Type: ApplicationFiled: January 22, 2013Publication date: July 24, 2014Applicant: MICROSOFT CORPORATIONInventors: Ruihua Song, Yu Chen, Ji-Rong Wen
-
Publication number: 20140207799Abstract: Various embodiments select features from a feature space. In one embodiment a candidate feature set of k? features is selected from at least one set of features based on maximum relevancy and minimum redundancy (MRMR) criteria. A target feature set of k features is identified from the candidate feature set, where k?>k. Each a plurality of features in the target feature set is iteratively updated with each of a plurality of k??k features from the candidate feature set. The feature from the plurality of k??k features is maintained in the target feature set, for at least one iterative update, based on a current MRMR score of the target feature set satisfying a threshold. The target feature set is stored as a top-k feature set of the at least one set of features after a given number of iterative updates.Type: ApplicationFiled: January 21, 2013Publication date: July 24, 2014Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATIONInventors: David HAWS, Dan HE, Laxmi P. PARIDA
-
Publication number: 20140207801Abstract: Disclosed are methods, apparatus, systems, and computer-readable storage media for recommending an event to a user. In some implementations, one or more servers receive information identifying a plurality of events. The one or more servers store data of the plurality of events in a first one or more data tables having an action field, an item field, and a user field, and analyze the data of the first one or more data tables to generate one or more pairs, each pair including information identifying a set of events and a target event. The one or more servers may calculate a similarity score for each of the one or more pairs and store the respective similarity score in a second one or more data tables having a set field, a target event field, and a similarity score field.Type: ApplicationFiled: January 21, 2014Publication date: July 24, 2014Applicant: SALESFORCE.COM, INC.Inventors: Joel Palmert, Tony Meng
-
Patent number: 8788495Abstract: Emotion data is added to tags. The emotion data is generated by a user upon accessing content. The emotion data is obtained from measurable physiological parameters that reflect the user's emotions while accessing the content. The emotion data can be used to provide services.Type: GrantFiled: March 30, 2010Date of Patent: July 22, 2014Assignee: International Business Machines CorporationInventor: Fang Rui Shu
-
Patent number: 8788340Abstract: In one embodiment, a social-networking system accesses a social graph, identifies a first set of concept nodes associated with an application, connected to friends of a user, selects a second set of concept nodes from the first set of concept nodes based on edges connected to the concept nodes, and generates an advertisement for the application, where the advertisement includes information representing the concept nodes in the second set of concept nodes.Type: GrantFiled: December 16, 2011Date of Patent: July 22, 2014Assignee: Facebook, Inc.Inventors: Alexander Paul Schultz, Thomas Alison
-
Patent number: 8788524Abstract: A query in an imprecise syntax is received, and a plurality of assumptions relating to the meaning of the query are determined. One of the assumptions is chosen from the determined plurality of assumptions, and a first response to the query is generated based on the chosen assumption. A user interface mechanism is provided to permit a user to change the assumption on which the first response was based, and a second response to the query is generated based on the changed assumption.Type: GrantFiled: May 14, 2010Date of Patent: July 22, 2014Assignee: Wolfram Alpha LLCInventors: Stephen Wolfram, Oyvind Tafjord
-
Patent number: 8788491Abstract: A hierarchical decoder is operable to decode a hierarchical multi-layer data package. The hierarchical multi-layer data package includes a plurality of layers and one or more subpackages in each layer. To decode the data package, an objective function is created based on parameters for identifying data in the data package most relevant to information of interest. Subpackages in one or more layers are evaluated using the objective function and a subpackage is selected. Information for the selected subpackage, which may include a summary and metadata, is used to determine whether the subpackage satisfies an information need. If the information need is not satisfied, a drill down procedure is performed to evaluate subpackages in a next lower layer of the data package.Type: GrantFiled: June 11, 2012Date of Patent: July 22, 2014Assignee: General Instrument CorporationInventor: Tom Tirpak
-
Publication number: 20140201219Abstract: The present disclosure relates to a method and apparatus for searching an image, and to a computer-readable recording medium for executing the method. The apparatus for searching an image of the present disclosure obtains features of an input image; and obtains words that correspond to the features respectively and an adjacent word that is adjacent to the words corresponding to the features. When a word is assigned to a first word cell of a plurality of word cells that are included in a visual feature space, an adjacent word is assigned to at least one second word cell that is adjacent to the first word cell, where the plurality of word cells is assigned to different words, and at least one word being within a predetermined distance from a word is designated as the adjacent word.Type: ApplicationFiled: November 12, 2012Publication date: July 17, 2014Inventors: Tae-Hoon Kim, Minje Park, Songki Choi
-
Publication number: 20140201173Abstract: Example embodiments relate to file-based social recommendations. In example embodiments, a system may search a social network to identify uploaded files that are similar to a file from a user and then identify unconnected users that uploaded the uploaded files, where the unconnected users are not connected to the user in the social network. In response to identifying the unconnected users, the system may then communicate a user recommendation that at least one unconnected users be added to the social context.Type: ApplicationFiled: January 15, 2013Publication date: July 17, 2014Applicant: HEWLETT-PACKARD DEVELOPMENT COMPANY, L.P.Inventors: Eyal Roth, Olga Tubman, Boris Korenfeld
-
Patent number: 8782067Abstract: In a searching device, an extractor extracts extracted documents, each of which contains a desired search string, from a plurality of document groups each including a plurality of documents. A preference determiner determines an output preference for each of the extracted documents in accordance with a predetermined rule. A group determiner determines, in sequence for each of the plurality of document groups, whether or not the document group is a determined group which contains one or more satisfying documents among extracted documents satisfying a predetermined output condition. And, an outputter outputs, in sequence for each of the plurality of the determined groups, a predetermined number of the satisfying documents contained in the determined group.Type: GrantFiled: March 23, 2012Date of Patent: July 15, 2014Assignee: Casio Computer Co., LtdInventor: Hiroyasu Ide
-
Patent number: 8782058Abstract: Various embodiments are described herein that generally relate to systems and methods for generating context specific terms and performing various actions based on the context specific terms. One example embodiment includes a computer-implemented method for generating context specific terms comprising obtaining a collection of terms from at least one electronic file associated with a given context; comparing the collection of terms with a collection of expected terms to generate candidate terms that are not in the collection of expected terms; determining a relevance for each of the candidate terms; and determining whether to add a given candidate term to a collection of context specific terms for the given context if the relevance for the given candidate term is above a threshold.Type: GrantFiled: October 12, 2011Date of Patent: July 15, 2014Assignee: Desire2Learn IncorporatedInventors: Brian Pearson, Jeremy Auger
-
Publication number: 20140195549Abstract: Disclosed in some examples are methods, systems and machine readable medium for recommending an out-of-network communication by determining a set of potential recommended members of a social networking service based upon one or more recommendation criteria. In some examples the recommendation criteria may include: a profile similarity to a previous target of an out-of-network communication, a degree of correspondence between an interest and intent of the sending member, and a likelihood of response.Type: ApplicationFiled: February 21, 2013Publication date: July 10, 2014Inventors: Junghoon Ahn, Abhishek Gupta, Christian Posse, Anmol Bhasin, Yurong Shi, Jian Li, Jacob Kessler
-
Publication number: 20140195548Abstract: Methods and systems to identify video content based on video fingerprint matching are described. In some example embodiments, the methods and systems generate a query fingerprint of a frame of video content captured at a client device, query a database of reference fingerprints, determine the query fingerprint of the frame of captured video content matches a reference fingerprint, and identify the video content based on the match of fingerprints.Type: ApplicationFiled: January 7, 2013Publication date: July 10, 2014Inventor: Wilson Harron
-
Patent number: 8775442Abstract: Techniques for providing semantic search of a data store are disclosed. A similarity metric of a document comprising the data store to a concept represented in a semantic model derived at least in part from a reference source that includes content not included in the data store is determined. A relevance metric of a search query to the concept is computed. The similarity metric and the relevance metric are used to determine, at least in part, a ranking of the document with respect to the search query.Type: GrantFiled: May 15, 2012Date of Patent: July 8, 2014Assignee: Apple Inc.Inventors: Jennifer Lauren Moore, Devang K. Naik, Jerome R. Bellegarda, Kevin Bartlett Aitken, Kim E. Silverman