Abstract: There is provided with an image searching apparatus. A searching unit searches a plurality of search target images for an image relating to a query image based on a first similarity to the query image by a first evaluation scheme. A first obtaining unit obtains a feedback indicating correct or incorrect regarding the searched image. A second obtaining unit obtains an evaluation of an effect of changing a similarity evaluation scheme to a second evaluation scheme from the first evaluation scheme in accordance with the feedback. A control unit controls, in accordance with the evaluation of the effect, a search of the plurality of search target images for the image relating to the query image based on a second similarity to the query image by the second evaluation scheme.

Abstract: A computer-implemented method according to one embodiment includes receiving a question in a natural language format; deriving a plurality of subqueries from the question; determining that a first subquery of the plurality of subqueries cannot be resolved utilizing a curated knowledge base; resolving the first subquery, utilizing unstructured text; and adjusting the curated knowledge base, utilizing the resolved first subquery.

Abstract: Provided is a system and method for automated feature engineering of multidimensional data. The system and method may generate a flat data set for use in training a machine learning model from a multidimensional data set. In one example, the method may include extracting metadata of multidimensional data, generating declarative statements of features that are to be derived from the multidimensional data based on hierarchical attributes of the multidimensional data which are identified from the extracted metadata, querying, via a query engine, the multidimensional data based on the generated declarative statements to generate the derived features, and storing the derived features in a training data set.

Abstract: A system includes a narrative repository which stores a plurality of narratives and, for each narrative, a corresponding outcome. A narrative evaluator receives the plurality of narratives and the outcome for each narrative. For each received narrative, a subset of the narrative is determined to retain based on rules. For each determined subset, a entropy matrix is determined which includes, for each word in the subset, a measure associated with whether the word is expected to appear in a sentence with another word in the subset. For each entropy matrix, a distance matrix is determined which includes, for each word in the subset, a numerical representation of a difference in meaning of the word and another word. Using one or more distance matrix(es), a first threshold distance is determined for a first word of the subset. The first word and first threshold are stored as a first word-threshold pair associated with the first outcome.

Abstract: A search system for providing a vertical service connection includes a search word completing unit configured to determine, upon receiving a search word from a user terminal, at least one auto-complete search word corresponding to a currently input search word; a search word matching unit configured to match the at least one auto-complete search word with a vertical service; a display unit configured to cause the display of the at least one auto-complete search word and a vertical service icon; and a vertical service searching unit configured to provide, upon receiving a selection signal indicating selection of the vertical service icon, results of searching content included in the matched vertical service for the at least one auto-complete search word.

Abstract: A data modeling platform allows a user to enter a search request using natural language (e.g., typed or spoken). The search request may be semantically parsed to provide mapping to various database search elements. Pre-existing data models may be identified from the database search elements. One or more new data models may be created from the database search elements. The results represent candidate data models in response to the user's search request.

Abstract: A system for and method of providing reusable software service information based on natural language queries. The system and method may include receiving, from a user system, query data in a natural language format that indicates a request for a plurality of reusable software service applications that are configured to perform a particular task, processing the query data to generate search criteria that include query values, and searching, a database, for the plurality of reusable software service applications based on the query values.

Abstract: Methods and apparatus include presenting an initial set of names to a user. The user selects a set of names from those presented. An Interactive Evolutionary Algorithm (IEA) extracts features of each selected name from a database of names and features to form a feature set. The IEA forms a set of match features that are chosen from the feature set according to a priority function and/or weighting of the features, either of which may vary in succeeding iterations. The IEA searches the database to obtain a candidate set of names, where each name has features matching the match features. One or more names is chosen from the candidate set and added into a presentation set of names. The IEA may repeat the formation of the match features, candidate set, and selection of one or more names from the candidate set until the new presentation set is complete.

Abstract: An approach for managing a set of data tuples is provided. A search tuple is received and a comparison relation is determined. The set of data tuples is ordered based on the comparison relation. A placement of the search tuple is determined between two consecutive data tuples in the ordered set of data tuples so that the search tuple is ordered among the ordered set of data tuples in accordance with the comparison relation. A presentation is generated based on the placement of the search tuple between the two consecutive data tuples. The presentation includes a sequence of data tuples included in the ordered set of data tuples. The sequence includes the two consecutive data tuples and the search tuple between the two consecutive data tuples.

Abstract: Natural language query translation may be provided. A statistical model may be trained to detect domains according to a plurality of query click log data. Upon receiving a natural language query, the statistical model may be used to translate the natural language query into an action. The action may then be performed and at least one result associated with performing the action may be provided.

Abstract: A computing device maintains an input history in memory. This input history includes input strings that have been previously entered into the computing device. When the user begins entering characters of an input string, a predictive input engine is activated. The predictive input engine receives the input string and the input history to generate a candidate list of predictive inputs which are presented to the user. The user can select one of the inputs from the list, or otherwise continue entering characters. The computing device generates the candidate list by combining frequency and recency information of the matching strings from the input history. Additionally, the candidate list can be manipulated to present a variety of candidates. By using a combination of frequency, recency and variety, a favorable user experience is provided.

Abstract: Embodiments disclosed herein may relate to identifying music from a text source utilizing a computing platform in a communication system.

Abstract: Embodiments of the invention provide a method and computer program products for information retrieval from multiple documents by proximity searching for search queries. A method includes generating an index for the multiple documents, wherein the index includes words in snippets in the documents. An input search query is processed against the index by searching query terms over the snippets to introduce term proximity information implicitly in the information retrieval. Results of multiple sentence level search operations are combined as output.

Abstract: A method can output content of interest of a structured electronic document from a computer or distributed computer system having a processor and memory. The method includes loading a common expression and a data structure definition into memory, the common expression identifying a content element in a first structured electronic document, the data structure definition defined according to the common expression; creating in memory an instance of a data structure defined by the data structure definition; applying with the processor the common expression to a second structured electronic document to extract a content element from the second structured electronic document; storing the extracted content element in the instance of the data structure; and populating a template structured electronic document using the instance of the data structure to produce an output structured electronic document.

Abstract: A question is received to be answered by a question answering (QA) system. The question may be a business intelligence question that is expressed in a natural language. The question is parsed. The parsed question is matched to a pattern from a number of patterns. A technical query associated with the matched pattern is processed to retrieve data relevant to the question from a number of data sources. The QA system generates an answer to the question based on retrieved data. In one aspect, the QA system generates answers based contextual information.

Abstract: A system may include multiple personal data sources and a machine-implemented data extractor and correlator configured to retrieve personal data from at least one of the personal data sources. The data extractor and correlator may extract information from unstructured data within the retrieved personal data and correlate the extracted information with previously stored structured data to generate additional structured data. The system may also include a storage device configured to store the previously stored structured data and the additional structured data. A natural language query module may be configured to receive a natural language query from a user and provide a response to the natural language query based at least in part on one or both of the previously stored structured data and the additional structured data.

Abstract: Information may be presented to a user by using a first application to receive a character stream of one or more noncompletion characters. The noncompletion characters indicate that additional characters may be received. The character stream is exchanged with a host that analyzes the character stream to generate results that are responsive to the user's predicted interest. Results are received and displayed so that the user may select one of the results to launch a code segment related to the selected result.

Abstract: A user interface application is created to prompt a user for information and to generate a query to be submitted to a natural language (NL) query answering system. A first user is prompted to submit an NL query, and the NL query is received. The user interface application is generated using the NL query and so that the user interface application is configured to prompt the first user or a second user to provide information, receive the information, generate a computer-generated query based on the NL query and the received information, provide the computer-generated query to the NL query answering system, receive answer information from the NL query answering system in response to the computer-generated query, and cause at least some of the answer information to be displayed on a display device.

Abstract: The current application is directed to characterization of e-commerce-related searches and digital encoding of the characterizations in a database or other data-storage system by an e-commerce search engine. As one example, when a user enters a query term in a search textbox, the user is provided with a choice of pressing an action button labeled “I want to buy” or an action button labeled “I want to sell,” referred to as a “buy button” and “sell button,” respectively. The search engine records the buy-button and sell-button inputs along with corresponding search queries. The search-engine implementation then collects these queries from a large number of users into a database and matches them over a period of time, thus helping buyers find sellers and vice versa.

Abstract: Systems and methods are described for identifying item attributes that are likely to be useful to users in refining searches for items. In one embodiment, when a user submits a search query to search for items, all or a portion of the search query is used to look-up a previously-generated list of item attributes (and optionally associated attribute values) that are associated with the search query and are useful for refining the search. These item attributes, and any associated attribute values, are then presented to the user in a selectable form such that the user can interactively narrow the search.

Abstract: A social networking system receives a query associated with a user and, in response, provides a combined result set comprising objects stored by a social networking system that match the query. The combined result set comprises multiple result sets obtained from different search algorithms. The various objects stored by the social networking system may be of different types representing different concepts, such as user objects, application objects, event objects, location objects, group objects, and hub/page objects, any of which may be included in the result set. The objects of the result set may be further filtered, ordered, and/or grouped based at least in part on known relationships of the user with the objects, such as geographic distances between locations associated with the user and the objects.

Abstract: A natural language query tool comprising cascaded conditional random fields (CRFs) (e.g., a linear-chain CRF and a skip-chain CRF applied sequentially) processes natural language input to produce output that can be used in database searches. For example, cascaded CRFs extract entities from natural language input that correspond to column names or column values in a database, and identify relationships between the extracted entities. A search engine can execute queries based on output from the cascaded CRFs over an inverted index of a database, which can be based on one or more materialized views of the database. Results can be sorted (e.g., according to relevance scores) and presented in a user interface.

Abstract: An exchange that fulfills natural language travel requests is described herein. According to some embodiments, systems for managing an exchange that fulfills natural language travel requests may include a publishing module that receives a plurality of inventory records from a plurality of suppliers, each of the plurality of inventory records pertaining to an available offering, the offering being defined by a set of metadata attributes, and a pattern recognition artificial intelligence engine that generates equivalent phrases for the set of metadata attributes from each inventory record by applying pattern recognition artificial intelligence to the set of metadata attributes, the equivalent phrases representing possible natural language queries to which an inventory record may correspond.

Abstract: Systems and methods for searching databases by sound data input are provided herein. A service provider may have a need to make their database(s) searchable through search technology. However, the service provider may not have the resources to implement such search technology. The search technology may allow for search queries using sound data input. The technology described herein provides a solution addressing the service provider's need, by giving a search technology that furnishes search results in a fast, accurate manner. In further embodiments, systems and methods to monetize those search results are also described herein.

Abstract: A method and system for searching information using information device is provided. The method for searching information comprises generating one or more questions corresponding to a search query. The questions are generated based on a predetermined ordering criterion when a predefined condition is true. One or more responses to the questions are received from a user. Information is rendered on the information device based on one or more of a response and the search query.

Abstract: Media and methods are provided for creating and operating a keyword expression language. Syntax is generated as an abbreviation to represent a list of keywords. The syntax is executed as part of the keyword expression language to provide keywords. The syntax includes tokens that substitute for groups of information. Advertisers generate syntax which is subsequently used by a third-party to match to search queries and ads. The third-party may also generate keywords to match to the search queries and ads. The keywords are used to trigger advertising over the Internet.

Abstract: A server system receives, respectively, a first character string from a first user and a second character string from a second user. There are one or more differences between the first and second character strings. The server system obtains from a plurality of previously submitted complete queries, respectively, a first set of predicted complete queries corresponding to the first character string and a second set of predicted complete queries corresponding to the second character string. There are one or more identical queries in both the first and second sets. The server system conveys at least a first subset of the first set to the first user and at least a second subset of the second set to the second user. Both the first subset and the second subset include a respective identical query.

Abstract: Computer-readable media, computer systems, and computing methods are provided for classifying search results as either of good quality or of poor quality. Initially, a portion of the search results, such as the highest ranked documents, are selected for evaluation. A level of quality for each of the selected search results is determined using a classification process that includes the following steps: targeting features demonstrated by the selected search results to be evaluated; evaluating the selected features to generate a level-of-quality score for each of the selected search results; comparing the score against a predefined threshold value; and, based on the comparison, assigning each of the selected search results an absolute measurement. The absolute measurement indicates poor quality when the score is less than the threshold value. Upon recognizing that the selected search results are of poor quality, a corrective action that reformulates the issued search query is automatically invoked.

Abstract: An improved search engine creates correlations linking terms from inputs provided by a user to selected target terms. The correlation search process receives pre-processed inputs from a user including a wide variety of input formats including keywords, phrases, sentences, concepts, compound queries, complex queries and orthogonal queries. The pre-processing also includes pre-processing of general digital information objects and static or dynamic generation of questions. After a correlation search of the information presented by the pre-processing, the search results are processed in novel ways to provide an improved relevance ranking of results.

Abstract: A system and method for representing, storing and retrieving real-world knowledge on a computer or network of computers is disclosed. Knowledge is broken down into permanent atomic “facts” which can be stored in a standard relational database and processed very efficiently. It also provides for the efficient querying of a knowledge base, efficient inference of new knowledge and translation into and out of natural language. Queries can also be processed with full natural language explanations of where the answers came from. The method can also be used in a distributed fashion enabling the system to be a large network of computers and the technology can be integrated into a web browser adding to the browser's functionality.

Abstract: In general, the subject matter described in this specification can be embodied in methods, systems, and program products for receiving user input that defines a search query, and providing the search query to a server system. Information that a search engine system determined was responsive to the search query is received at a computing device. The computing device is identified as in a first state, and a first output mode for audibly outputting at least a portion of the information is selected. The first output mode is selected from a collection of the first output mode and a second output mode. The second output mode is selected in response to the computing device being in a second state and is for visually outputting at least the portion of the information and not audibly outputting the at least portion of the information. At least the portion of information is audibly output.

Abstract: A paleogeographic search system is provided for locating geoscience data relevant to a geographic search aperture. The paleogeographic search system comprises a geodatabase, an interface, a paleogeographic reconstruction engine, and a host processor. The paleogeographic engine is configured to transform the geographic search aperture according to tectonic plate movements to a transformed search aperture that the geographic search aperture occupied in a geological age of interest. The host processor is configured to apply a paleoaugmented geographic search aperture to a geodatabase for additional oil exploration data relevant to the geographic search aperture today.

Abstract: A system for creating keywords or phrases to be used for mining online information has a text source accessible by the system, a keyword aggregation function, and a mechanism for configuring the system. The system extracts keywords and phrases from the text source according to the configuration, for submission to a search facility.

Abstract: This invention provides a query and note based content creation, sharing and search system that allows users to create and manage notes, to share the notes with others, and to query the notes from a cellphone or computer interface. This is different from the traditional Internet and search engines in which user queries can only be indirectly answered. This invention allows users to query the notes from a cellphone or a computer when a prompt and precise response is needed. This invention also allows one or more programs to be specified for a note so when the note is returned to the user the program(s) can be executed.

Abstract: A system for and method of providing reusable software service information based on natural language queries. The system and method may include receiving, from a user system, query data in a natural language format that indicates a request for a plurality of reusable software service applications that are configured to perform a particular task, processing the query data to generate search criteria that include query values, and searching, a database, for the plurality of reusable software service applications based on the query values.

Abstract: Techniques for utilizing data mining technology to extract universal topics with multilingual representations from a multilingual database, and to organize existing or new documents in different languages by analyzing their respective topic distributions.

Abstract: An online address book system having sufficient hardware and software to operate an address book user interface and to perform intelligent interpretations of voice and text inputs from users. The system includes at least one server software module that includes software to perform a plurality of functions. These include the ability to receive voice input data and separate user voice queries, wherein the software can arrange the data so as to create a data base that includes at least three access dimensions, including contact access, contact-relationship access and contact-time frame access, and so as to create a connectivity matrix based on a plurality of contact pair relationships applying connective recognition logic. The system provides a voice operated user interface that permits access to address book stored data based on user input selected from the group consisting of contact, a contact-relationship pair, a contact-time frame pair, and combinations thereof.

Abstract: Various methods for computing relevancy are provided. One example method includes determining a behavior relevance factor for a user interface item, determining a semantic relevance factor for the user interface item, and determining an internal state of a user relevance factor for the user interface item. The example method further includes generating an overall relevance value for the user interface item. The overall relevance value may be a function of the behavior relevance factor, the semantic relevance factor, and the internal state of the user relevance factor. Similar and related example methods and example apparatuses are also provided.

Abstract: A semantic search system using a semantic ranking scheme including: an ontology analyzer analyzing ontology data related to a search target to determine a weight value of each property according to a weighing method for property; a semantic path extractor extracting all the semantic paths between resources and query keywords and determining a weight value of each extracted semantic path according to the semantic path weight value determination scheme by using the weight value of each property; a relevant resource searcher traversing an instance graph of ontology based on a semantic path having a pre-set length and weight value of more than an expectation level to search resources that have a semantic relationship with the query keywords and are declared as a type presented in the query; and a semantic relevance ranker selecting a top-k results having the highest rank from among the candidate results extracted by the relevant resource researcher by using a relevance scoring function.

Abstract: Semantic queries are expressed and executed within a relational database. This can be done by defining semantic rules applied to execute the semantic queries using table valued functions and common table expressions, and then simply calling the defined table valued functions to execute the queries.

Abstract: A tag is received from a member of a social network site. The tag is associated with a favorite of the member. Tag data obtained from the tag is provided to a second site that hosts a searching service. The second site determines the relevancy of the tag data relative to a new search request received by a user on the second site.

Abstract: Systems, methods, and computer-storage media for generating contextual queries are provided. The system includes a search engine to receive user queries and contexts, a query understanding component to generate a semantic representation of the query, and a data source command generator to transform the semantic representation into commands for multiple data sources. The data source command generator is connected to the query understanding component. The data source command generator selects data source commands based on lexical information associated with each data source.

Abstract: Embodiments of the invention broadly contemplate systems, methods, apparatuses and program products providing a faceted search tool enabling a user to enter a natural language query that contains representative terms the user wants to search for, and a convenient interface display in which the facets which contain the searched values are automatically resolved and presented.

Abstract: Computer-readable media and a computing device are described for providing geotemporal search and a search interface therefor. A search interface having a location portion and a timeline portion is provided. A geographic area is selected in the location portion by adjusting the visible area of a map. A temporal window is selected in the timeline portion by adjusting sliders along a timeline to a desired start and end time. The start and end times can be in the past, present, or future. A geotemporal search is executed based on the selected geographic area and temporal window to identify search results having associated metadata indicating a relationship to the selected geographic area and temporal window. One or more search terms are optionally provided to further refine the geotemporal search.

Abstract: Disclosed herein are systems, methods, and computer-readable storage media for providing an N-best list interface. A system practicing the method receives a search query formatted according to a standard language for containing and annotating interpretations of user input, the search query being based on a natural language spoken query from a user and retrieves an N-best list of recognition results based on the search query. The system then transmits the N-best list of recognition results to a user device, receives multimodal disambiguation input from the user, the input indicating an entry in the N-best list, and transmits to the user device additional information associated with the selected entry. The additional information can be a map indicating an address for the selected entry. The standard language can be XML-based Extensible MultiModal Annotation (EMMA) markup language from W3C.

Abstract: Disclosed herein are systems, methods, and computer-readable storage media for an iterative disambiguation interface. A system practicing the method receives a search query formatted according to a standard XML markup language for containing and annotating interpretations of user input, the search query being based on a natural language spoken query from a user and retrieves search results based on the search query. The system transmits the search results to a user device and iteratively receives multimodal input from the user to change search attributes and transmits updated search results to the user device based on the changed search attributes. The search results can include a link to additional information, such as a video presentation, related to the search results. The standard XML markup language can be Extensible MultiModal Annotation (EMMA) markup language from W3C. The system can generate an iteration transaction history for each multimodal input and updated search result.

Abstract: Disclosed herein is a natural language-based tour destination recommendation apparatus and method. The natural language-based tour destination recommendation apparatus includes a query analysis unit, a tour destination search unit, and a tour destination recommendation and provision unit. The query analysis unit performs linguistic analysis on a user's tour-related query and then extracting query analysis information to be used for figuring out the user's intention from a document index DB. The tour destination search unit searches a tour destination DB for one or more recommended tour destinations using the extracted query analysis information. The tour destination recommendation and provision unit provides the retrieved recommended tour destinations to the user.

Abstract: A method for intent mining is provided. The method includes performing a preliminary search of a constrained source using one or more seed phrases to generate multiple preliminary search results representing different ways of expressing a desired intent. The method also includes identifying each of the plurality of preliminary search results that have expressed the desired intent to generate a plurality of intent results. The method also includes producing multiple action search strings around one or more action verbs in each of the multiple intent results. The method further includes applying each of the multiple action search strings on one or more non-constrained sources to generate multiple action search results.

Abstract: A search system will receive a voice query and use speech recognition with a predefined vocabulary to generate a textual transcription of the voice query. Queries are sent to a text search engine, retrieving multiple web page results for each of these initial text queries. The collection of the keywords is extracted from the resulting web pages and is phonetically indexed to form a voice query dependent and phonetically searchable index database. Finally, a phonetically-based voice search engine is used to search the original voice query against the voice query dependent and phonetically searchable index database to find the keywords and/or key phrases that best match what was originally spoken. The keywords and/or key phrases that best match what was originally spoken are then used as a final text query for a search engine. Search results from the final text query are then presented to the user.

Abstract: A method for searching a database of digital media assets, comprising: designating a database of digital media assets, wherein the database of digital media assets has been indexed according to a set of general indexers; receiving a search query; defining specialized search conditions by identifying one or more elements of the search query corresponding to one or more specialized indexers; defining general search conditions by identifying elements of the search query corresponding to the general indexers; identifying a subset of the digital media assets by applying the general search conditions; indexing the subset of the digital media assets using the identified specialized indexers; and ranking the subset of the digital media assets by applying the specialized search conditions.