Abstract: A method for assisting query comprises acquiring a record that includes a unique identification and at least one piece of dimension information associated with the unique identification, wherein each piece of dimension information includes a sequence of attributes, each attribute being ordered in the sequence in accordance with a hierarchically progressive relation; calculating a hash value for an attribute pair comprising attributes of every two neighboring levels; and storing, in association with the hash value, the unique identification in at least one storage device. A method for querying comprises receiving a query; calculating one or more hash values; in response to the one or more hash values matching stored hash values in at least one storage device; deriving an intersection of unique identifications that are stored in association with the matched stored hash values; and returning at least one unique identification included in the intersection as a query result.

Abstract: In one aspect, a device includes a processor and a memory accessible to the processor. The memory bears instructions executable by the processor to access past messages associated with the device, and determine that at least a first multi-term phrase appears in plural messages among the past messages. The instructions are also executable by the processor to, responsive to input invoking message composition and based at least in part on a determination that the multi-term phrase appears in plural messages among the past messages, present on the device at least plural terms from the multi-term phrase in an order in which the at least plural terms appear in the multi-term phrase as a candidate message.

Abstract: A method for identifying similarity between query samples and stored samples in an efficiently maintained reference library may include receiving a first threshold and a second threshold, receiving a plurality of binary reference samples, and processing each reference sample of the plurality of reference samples. The processing may include operations of assigning each reference sample a respective unique identifier, producing a reference sample fingerprint for each reference sample, and registering each respective unique identifier to reference sample fingerprint pair in a reference library.

Abstract: A system, method, and computer-readable medium, is described that finds similarities among programming applications based on semantic anchors found within the source code of such applications. The semantic anchors may be API calls, such as Java's package and class calls of the JDK. Latent Semantic Indexing may be used to process the application and semantic anchor data and automatically develop a similarity matrix that contains numbers representing the similarity of one program to another.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media for speech recognition. One method includes obtaining an input acoustic sequence, the input acoustic sequence representing an utterance, and the input acoustic sequence comprising a respective acoustic feature representation at each of a first number of time steps; processing the input acoustic sequence using a first neural network to convert the input acoustic sequence into an alternative representation for the input acoustic sequence; processing the alternative representation for the input acoustic sequence using an attention-based Recurrent Neural Network (RNN) to generate, for each position in an output sequence order, a set of substring scores that includes a respective substring score for each substring in a set of substrings; and generating a sequence of substrings that represent a transcription of the utterance.

Abstract: This application discloses a method implemented on a client system for creating an automatic reply (auto-reply) message. The client system obtains content information of a webpage including information concerning a plurality of content items. A graphical user interface is displayed and includes a preview of the webpage including the plurality of content items of the webpage. Then, the auto-reply message is composed in accordance with a user selection of a specific content item of the webpage in the webpage preview, and at least includes corresponding information of the selected content item. The auto-reply message is further associated with one or more predefined keywords according to one or more auto-reply rules, and sent with these predefined keywords to a server system. The server system is configured to return the auto-reply message automatically to a subscriber device in response to a user inquiry including at least one of the predefined keywords.

Abstract: There is provide a computing system configured to receive an object identifier and contextual information from an end-user, to compute the object identifier and contextual information based on pre-defined set of rules, to map said object identifier and contextual information to an entry point associated with a specific computer application among a plurality of applications, and to provide access to said specific computer application to said end-user. There is also provided a method to do the same, and a tag for use with a physical object, the tag comprising a redirection identifier embodying a unique identifier of the object and an HTTP address of a server permitting to redirect or provide access to the user to a specific computer application about the object.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for language modeling of complete language sequences. Training data indicating language sequences is accessed, and counts for a number of times each language sequence occurs in the training data are determined. A proper subset of the language sequences is selected, and a first component of a language model is trained. The first component includes first probability data for assigning scores to the selected language sequences. A second component of the language model is trained based on the training data, where the second component includes second probability data for assigning scores to language sequences that are not included in the selected language sequences. Adjustment data that normalizes the second probability data with respect to the first probability data is generated, and the first component, the second component, and the adjustment data are stored.

Abstract: In order to solve a conventional problem that an accurate translation cannot be realized, a learning apparatus includes: a parser unit parsing a target language sentence, thereby acquiring a binary tree of the target language sentence; a source language element acquiring unit acquiring one or more source language elements; a source language partial structure acquiring unit acquiring one or more source language partial structures each containing a parent node having a phrase label and two child nodes each having a phrase label or a source language element; a labeling unit providing a reordering label to the one or more source language partial structures; a model building unit that builds one or more parsing models each having appearance probability information regarding a labeled source language partial structure; and an accumulating unit accumulating a binary tree of a source language sentence having the one or more parsing models.

Abstract: A server is configured to receive, from a client, a query and context information associated with a document; obtain search results, based on the query, that identify documents relevant to the query; analyze the context information to identify content; generate first scores for a hierarchy of topics, that correspond to measures of relevance of the topics to the content; select a topic that is most relevant to the context information when the topic is associated with a greatest first score; generate second scores for the search results that correspond to measures of relevance, of the search results, to the topic; select one or more of the search results as being most relevant to the topic when the search results are associated with one or more greatest second scores; generate a search result document that includes the selected search results; and send, to a client, the search result document.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for recognizing names of entities in speech. In one aspect, a method includes actions of receiving an utterance that includes (i) a first term that indicates a particular entity type, and (ii) a second term that indicates an entity name. Additional actions include obtaining a phonetic representation of the second term and determining that the phonetic representation of the second term matches a particular phonetic representation of a particular canonical name of a set of canonical names associated with a particular entity. Further actions include outputting a reference name associated with the particular entity as a transcription of the second term.

Abstract: A method, system and computer-usable medium for linking a set of executable code snippets to perform a complicated task, comprising: decomposing a natural language statement into a plurality of decomposed natural language components; searching a repository of code snippets to identify code snippets corresponding to each of the decomposed natural language components; ordering execution of the code snippets based upon the plurality of decomposed natural language components; and, executing the code snippets in order of the natural language statement requests until a final outcome is achieved.

Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for providing ranked search results responsive to a query. In one aspect, a method includes receiving, from a user device, a query specifying an entity, determining two or more resources each of which provide media content related to the entity, for each of the resources, determining an effectiveness measure that is a measure of the effectiveness of the resource to present, to a user of the user device, the media content related to the entity, ranking the resources using, at least in part, the respective effectiveness measure, and providing, to the user device, a presentation of search results for the ranked resources.

Abstract: Prosodic features are used for discriminating computer-directed speech from human-directed speech. Statistics and models describing energy/intensity patterns over time, speech/pause distributions, pitch patterns, vocal effort features, and speech segment duration patterns may be used for prosodic modeling. The prosodic features for at least a portion of an utterance are monitored over a period of time to determine a shape associated with the utterance. A score may be determined to assist in classifying the current utterance as human directed or computer directed without relying on knowledge of preceding utterances or utterances following the current utterance. Outside data may be used for training lexical addressee detection systems for the H-H-C scenario. H-C training data can be obtained from a single-user H-C collection and that H-H speech can be modeled using general conversational speech. H-C and H-H language models may also be adapted using interpolation with small amounts of matched H-H-C data.

Abstract: A method and device provide a word pair relevancy measure for a seed word-keyword pair. A set of documents is provided from which a co-occurrence matrix is created. The co-occurrence matrix is utilized to determine a transition probability and an expected search depth from which the word pair relevancy is determined.

Abstract: For language elements that indicate or suggest time, such as adverbs, these also contain date and time information that can be used to quantify time for a single piece of text or for an entire repository. This quantification of time can then be used by many applications, such as a mobile device that needs to know when to execute a command or when an investigator is trying to piece together a chain of events from different documents.

Abstract: In particular embodiments, one or more computer systems of a social-networking system determine available media content from one or more content sources and query a social graph of a social-networking system for social content associated with a user of the social-networking system. The social graph includes nodes and edges connecting the nodes. The nodes include user nodes that are each associated with a particular user of the social-networking system. The one or more computer systems of the social-networking system determine, using the social content and the available media content, a plurality of social coefficients for the user, determine, based on the determined social coefficients, recommended media content for the user, and provide the recommended media content for display in a programming guide on a display device.

Abstract: Disclosed are a word comfort/discomfort index prediction apparatus and method therefore. The word comfort/discomfort index prediction apparatus includes: a calculation unit calculating emotional associations between the word and one or more respective comparative words, calculating emotional associations between at least one predefined reference word and the one or more respective comparative words, and calculating an emotional similarity between the word and each of the at least one reference word; and a prediction unit predicting the comfort/discomfort index of the word by using the at least one emotional similarity and a predefined comfort/discomfort index of the at least one reference word.

Abstract: A word mapping method is provided, based on a plurality of connections between a plurality of words as generated by an individual or group, the map being formed based on connections between words provided by an individual or individuals. The resultant word map may be used for various purposes including cognitive diagnoses, sentence structure optimization of documents, and the like.

Abstract: An apparatus includes: a receiving unit configured to receive an input operation from a user performed on an accepting unit displayed on a display unit; a determination unit configured to determine whether or not detailed information set in the accepting unit is displayed on the display unit in accordance with the input operation received by the receiving unit, and at least one of a display state of the accepting unit and an input area of the input operation in the accepting unit; and a display control unit configured to control a display of the accepting unit displayed on the display unit in accordance with a determination result provided by the determination unit.

Abstract: A data transfer system includes a virtual printer driver to receive a report from a first system, the report in the form of a print job, and to generate an electronic document representative of the report; and an extraction module to extract a parameter from the electronic document, format the parameter in a first format, and provide the parameter in the first format to a second system.

Abstract: A system and method for automatically generating a narrative story receives data and information pertaining to a domain event. The received data and information and/or one or more derived features are then used to identify a plurality of angles for the narrative story. The plurality of angles is then filtered, for example through use of parameters that specify a focus for the narrative story, length of the narrative story, etc. Points associated with the filtered plurality of angles are then assembled and the narrative story is rendered using the filtered plurality of angles and the assembled points.

Abstract: Disclosed herein are systems, methods, and non-transitory computer-readable storage media for approximating responses to a user speech query in voice-enabled search based on metadata that include demographic features of the speaker. A system practicing the method recognizes received speech from a speaker to generate recognized speech, identifies metadata about the speaker from the received speech, and feeds the recognized speech and the metadata to a question-answering engine. Identifying the metadata about the speaker is based on voice characteristics of the received speech. The demographic features can include age, gender, socio-economic group, nationality, and/or region. The metadata identified about the speaker from the received speech can be combined with or override self-reported speaker demographic information.

Abstract: A graph syntax validation system, method, or computer-readable medium that receives: (i) an input graph, (ii) transformation rules, and (iii) a minimal valid graph. The system/method/computer-readable medium transforms the input graph into the minimal valid graph using the transformation rules that are comprised of source patterns and target patterns. The system/method/computer-readable medium recurrently transforms the input graph until either the input graph has been reduced to the minimal valid graph indicating that the input graph uses a valid syntax, or until it is determined that one or more transformation rules do not match the input graph indicating that the input graph uses an invalid syntax.

Abstract: The exemplary embodiments described herein are related to techniques for automatically generating narratives about data based on communication goal data structures that are associated with configurable content blocks. The use of such communication goal data structures facilitates modes of operation whereby narratives can be generated in real-time and/or interactive manners.

Abstract: The exemplary embodiments described herein are related to techniques for automatically generating narratives about data based on communication goal data structures that are associated with configurable content blocks. The use of such communication goal data structures facilitates modes of operation whereby narratives can be generated in real-time and/or interactive manners.

Abstract: Embodiments of the present invention provide information processing systems and methods for proliferating testing data based on sample testing data. In one embodiment, a difference is determined by comparing a desired query result to a query result acquired by executing a query statement on sample testing data. Sample testing data can then be proliferated based, at least in part, on the difference and data generation constraint conditions.

Abstract: Methods and system for generating a first set of selection options for presentation on a display screen of a television. Each first set selection option identifies a sub-set of symbols. The presentation provides a visual representation of spatial relationship between the first set of selection options and a first group of input options available on a physical remote control. A first user selection is received for the first group of input options. In response to receiving the first user selection, a second set of selection options corresponding to the first user selection, is generated for presentation as a visual representation of spatial relationship between the second set of selection options and a second group of input options available on the physical remote control. A second user selection is received for the input options from the second group and is interpreted to identify selection of one of the sub-set of symbols.

Abstract: Embodiments disclose a recommendation engine that implements a system and method for classifying users and media content, and for determining which media items a particular user will most likely enjoy. By organizing users and media content in a 3 dimensional virtual space and assigning x, y, and z coordinates to each, their relationship and affiliation with/to one another is defined by their distances from each other. When a user rates a media item highly, both that user and that media item move closer together. When a user gives a negative rating to a media item, they move farther apart. Further multiple users who rate the same media item similarly move closer together to form clusters of similar interest groups. This terraforms the virtual space into a natural attraction/separation of like-minded/different-minded users and similar/dissimilar media content.

Abstract: Information is obtained from the Internet is combined with processed and rated information from social networking services by a structured and social data aggregator providing highly relevant search results. In accordance with an embodiment of the present invention, there is provided a product information crawler, which crawls the Internet in search of web pages relevant to a requested subject or product. The product information crawler conveys captured web pages to a structured data extractor, which extracts product information. A social networking crawler crawls social networking services in search of social network information to the requested subject or product. The processed product information and processed social network information is conveyed to an information aggregator, which merges the information and stores it in a data store that can be queried by a user.

Abstract: Systems, methods and apparatuses for analyzing a string of terms (e.g., a search query, text of an email, and the like) are provided. In some examples, a determination is made as to whether one or more terms in the string matches a keyword. If so, various parts of speech of one or more terms in the string of terms may be determined. In some examples, a category of risk of the terms for which the part of speech is identified may also be determined. A risk rating may then be determined for the string of terms based on the relationship between the terms (e.g., the parts of speech) and the category or categories identified. In some examples, one or more additional actions may be implemented based on the risk rating.

Abstract: Systems and methods are provided for adding new nodes to a computer networked system. The systems and methods may identify a first set of nodes in a networked computer system. The first set of nodes may be included in a first hash computation that clusters the first set of nodes into communities. An application shard space including a first space and a second space may be generated. The first set of nodes may be mapped to application shards in the first space based on the first hash computation. The application shards in the first space may be assigned to a first set of machines of the networked computer system. The second space may be maintained for mappings of nodes not included in the first hash computation to application shards in the second space.

Abstract: The device includes a centralized warning system able to detect failures of systems of the aircraft and to display procedure lines on a screen, a procedure line including a text giving an indication of an action to be executed at least in the case of a failure, a semantic analysis module for determining from a current procedure line if the action to be executed is associated with a controllable control button and for implementing a semantic analysis of the text of the procedure line using word recognition and in order to derive therefrom indications relative to the controllable control button, these indications including at least an identification of the control button, and a warning unit receiving these indications and being able to warn a member of the crew.

Abstract: In one embodiment, a translator quality assurance system may rate a translation using a mapping chart. A data interface 150 may access a translation dictionary matching a source language to a target language. A processor 120 may map automatically a sentence pair having a source sentence 310 in a source language and a target sentence 320 in a target language at a linguistic level using the translation dictionary. A user output device 170 may display a translation quality report of the target sentence 320.

Abstract: The present invention provides a method and a system for processing a plurality of data feeds. The method and system include acquiring the plurality of data feeds from one or more content servers via a network, adding a respective task of a plurality of tasks to each of the plurality of data feeds, queuing the plurality of tasks at a queuing database, fetching each of the plurality of tasks queued at the queuing database by one or more worker servers and asynchronously processing the plurality of tasks by the one or more worker servers. Each respective task of the plurality of tasks is a request for processing a data feed of the plurality of data feeds.

Abstract: Methods and apparatus related to determining an edit rule based on a plurality of edits. Some implementations are directed to identifying the plurality of edits in one or more documents and determining an edit rule based on the pre-edit and post-edit phrases of the edits. Some implementations are directed to identifying the edits from one or more mature documents. The determined edit rule may be utilized to determine one or more candidate rephrasings of a subsequent phrase.

Abstract: An image output apparatus includes an image output unit, a presumption unit, a comparison unit and an output control unit. The image output unit outputs, in a predetermined format, an image of a content of a document file. The presumption unit presumes a first region/language setting of a document file created environment where the document file is created. The comparison unit compares the first region/language setting with a second region/language setting of the image output apparatus. The output control unit makes the image output unit output the image of the content of the document file by using a region/language setting selected from the region/language settings, which are compared with each other by the comparison unit, on the basis of a result of the comparison by the comparison unit in accordance with a predetermined selection condition.

Abstract: A method, a system, and computer readable medium comprising instructions for monitoring caller experience are provided. The method comprises obtaining caller experience of at least one step in a call flow, selecting a plurality of steps from the at least one step to form a group, presenting caller experience of the plurality of steps in first level of the call flow, and presenting the group as a step in a second level of the call flow.

Abstract: A method of providing an input method for an electronic device includes receiving a command for switching a display image of the electronic device; determining whether at least two words correspond to at least two languages when at least the two words are in the display image; determining a word count of a first language and a word count of a second language when the two words correspond to the two languages; and enabling a first input method corresponding to the first language when the word count of the first language is greater than the word count of the second language, or enabling a second input method corresponding to the second language when the word count of the second language is greater than the word count of the first language.

Abstract: Computer implemented techniques for performing transliteration of input text in a first character set to a second character set are disclosed. The techniques include receiving input text and determining a set of possible transliterations of the input text based on a plurality of mapping standards. Each mapping standard defines a mapping of characters in the first character set to characters in the second character set. The techniques further include determining a set of candidate words in the target language based on the possible transliterations and a text corpus. The techniques also include determining a likelihood score for each one of the candidate words based on a language model in the target language previously received words. The techniques also include providing one or more candidate words based on the likelihood scores and receiving a user selection indicating one of the candidate words.

Abstract: A natural language question answering system and method, and a paraphrase module are provided. The natural language question answering system includes a conversion module configured to generate a plurality of modified questions by paraphrasing a user's question; a plurality of question answering engines configured to receive each of the user's question and the modified questions, and select candidate answers corresponding to each of the user's question and the modified questions; and a detection module configured to detect at least one among the selected candidate answers as an answer.

Abstract: Various embodiments of technologies for information coding based on semantic data reasoning are provided. In some examples, a process performs semantic reasoning on semantic sentences of a textual message to generate a code, and embeds the code in the textual message as additional information.

Abstract: A system and method for organizing localization content for video game development is disclosed. The method includes generating executable instructions for a video game being developed, wherein the video game is being developed for deployment in a plurality of natural languages, wherein text strings and/or multimedia data to be rendered during game play are referenced by the executable instructions. The method further includes identifiably storing the text strings and/or multimedia data in one or more dictionary files such that the text strings and/or multimedia data are identifiably referenced by references in the executable instructions and are identifiably referenced by a corresponding natural language.

Abstract: A process for identifying, accurately selecting, and storing scientific data that is present in textual formats. The process includes providing scientific data located in a text document and searching the text document using a computer and selecting a plurality of key words and phrases using an algorithm. The selected key words and phrases are matched with a plurality of semantic definitions and a plurality of semantic definition-key words and phrase pairs are created. The created plurality of semantic definition-key words and phrase pairs are displayed to a user via a computer user interface and the user selects which of the created plurality of semantic definition-key words and phrase pairs are accurate. The process also includes storing the selected and accurate semantic definition-key words and phrase pairs in computer memory.

Abstract: A document is received that has a plurality of lines with text. This document includes text associated with at least one topic of interest and text not associated with the at least one topic of interest. Thereafter, it is determined, for each line in the document, a length of the line and a number of off-topic indicators with the off-topic indicators characterizing portions of the document as likely being not being associated with the at least one topic of interest. Thereafter, a density for each line can be determined based on the determined line length and the determined number of off-topic indicators. The determined densities for each line are used to identify portions of the documents likely associated with the at least one topic of interest so that data characterizing the identified portions of the document can be provided. Related apparatus, systems, techniques and articles are also described.

Abstract: Techniques for generating and providing phrases are described herein. These techniques may include analyzing one or more sources to generate a first corpus of phrases, each of the phrases for use as an identifier and/or for association with a user. Once a first corpus of phrases has been generated, these phrases may be filtered to define a second corpus of phrases. Phrases of this second corpus may then be suggested to one or more users. In some instances, the phrases suggested to a particular user are personalized to the user based on information previously known about the user or based on information provided by the user.

Abstract: Embodiments are provided for organization and presentation of content. In some embodiments, a plurality of images and a plurality of similarity rules for image categorization are received. For each image in the plurality of images, that image and each remaining image from the plurality is compared by: applying each similarity rule to the image and a remaining image from the plurality to obtain a numeric result, and recording the numeric result for the pair of images in a numeric representation, the numeric representation embodying similarities. The numeric representation is used as a reference for clustering the plurality of images into clusters of similar images, and each image is stored with a marker denoting a cluster to which it has been assigned.

Abstract: A method for adapting a language model for a context of a domain, comprising obtaining textual contents from a large source by a request directed to the context of the domain, discarding at least a part of the textual contents that contain textual terms determined as irrelevant to the context of the domain, thereby retaining, as retained data, at least a part of the textual contents that contain textual terms determined as relevant to the context of the domain, and adapting the language model by incorporating therein at least a part of the textual terms of the retained data, wherein the method is performed on an at least one computerized apparatus configured to perform the method and equipped for communication with the large source, and an apparatus for performing the same.

Abstract: A method for building a factual database of concepts and entities that are related to the concepts through a learning process. Training content (e.g., news articles, books) and a set of entities (e.g., Bill Clinton and Barack Obama) that are related to a concept (e.g., Presidents) is received. Groups of words that co-occur frequently in the textual content in conjunction with the entities are identified as templates. Templates may also be identified by analyzing parts-of-speech patterns of the templates. Entities that co-occur frequently in the textual content in conjunction with the templates are identified as additional related entities (e.g., Ronald Reagan and Richard Nixon). To eliminate erroneous results, the identified entities may be presented to a user who removes any false positives. The entities are then stored in association with the concept.

Abstract: A system and method for extracting facts from documents. A fact is extracted from a first document. The attribute and value of the fact extracted from the first document are used as a seed attribute-value pair. A second document containing the seed attribute-value pair is analyzed to determine a contextual pattern used in the second document. The contextual pattern is used to extract other attribute-value pairs from the second document. The extracted attributes and values are stored as facts.