97 results about "Semantic context" patented technology

A system and method for generating a frame of reference for a plurality of information, the plurality of information containing text data and obtained by a user through interaction with one or more information sources. The method and system include receiving selected information for analysis, the information including a plurality of text data and identifying a plurality of logical units of the text data. Also included are identifying a plurality of individual textual portions in each of the logical units and calculating the number of logical units associated with each of the individual textual portions of the plurality of textual portions for use in identifying a plurality of patterns including a respective pattern for each of the individual textual portions. Based on the identified patterns, a measure of importance for each of the plurality of patterns is calculated and respective patterns from the plurality of patterns are selected based on their corresponding measure of importance satisfying a predefined pattern importance threshold for retaining the respective patterns in the plurality of patterns. Further, a plurality of information context definitions is generated based on the selected respective patterns and the generated information context definitions are assigned as context definitions of a semantic context associated with the frame of reference. The plurality of information is related to each other and such relation is represented by the generated context definitions of the semantic context.

A method and system include sending a search context to a collaboration server for use in determining another user of the collaboration server that is associated with similar subject matter, the search context including at least one first context definition such that the first context definition has one or more words selected from a first plurality of information belonging to a first search information set of the user. The first plurality of information is related to each other and such relation is represented by the first context definition of the search context. The system and method also include receiving an identification of a matching search context associated with the another user such that the matching search context contains at least one second context definition considered to match the first context definition. The matching second context definition includes one or more words selected from a second plurality of information belonging to a second search information set associated with the another user.

Presented is a system and method for converting or translating expressions in a natural language such as English into machine executable expressions in a formal language. This translation enables a transformation from the syntactic structures of a natural language into effective algebraic forms for further exact processing. The invention utilizes algorithms employing a reduction of sequences of terms defined over an extensible lexicon into formal syntactic and semantic structures. This term reduction incorporates both syntactic type and semantic context to achieve an effective formal representation and interpretation of the meaning conveyed by any natural language expression.

A file includes data and semantic context. The semantic context identifies limitations on the use of the data in the file. Semantic context may include, among other possibilities, source information about a parameter, acceptable values for a parameter, presentation information, a record of the use of the data, and information to synchronize the use of the data with another file.

A computer-implemented method of training a neural network includes training a first neural network of a self organizing map type with a first set of first text documents each containing one or more keywords in a semantic context to map each document to a point in the self organizing map y semantic clustering; determining, for each keyword in the first set, all points in the self organizing map to which first documents containing said keyword are mapped, as a pattern and storing said pattern for said keyword in a pattern dictionary; forming at least one sequence of keywords from a second set of second text documents each containing one or more keywords in a semantic context; translating said at least one sequence of keywords into at least one sequence of patterns using the pattern dictionary; and training a second neural network with the at least one sequence of patterns.

A computerized interactor system uses physical, three-dimensional objects as metaphors for input of user intent to a computer system. When one or more interactors are engaged with a detection field, the detection field reads an identifier associated with the object and communicates the identifier to a computer system. The computer system determines the meaning of the interactor based upon its identifier and upon a semantic context in which the computer system is operating. One specific embodiment of the present invention is a bead interactor system that is a user playable sound and light show system. When an interactor bead is positioned within the detection space of the bead interactor system, a sound sequence begins and continues to play unaltered until the bead interactor is removed or other bead interactors are positioned within or removed from the detection space. Each bead represents a different sound and the row and column location of the bead within the detection space controls how the sound is modified, e.g., louder or softer, higher pitched or lower pitched, the period of play, etc. In some embodiments, the beads are translucent in order to conduct light from light sources located under each bead receptacle. The available sounds are determined not only by the identity of the beads and their positioning within the detection space and their associated states or orientations, but also by sound data stored in an accompanying computer readable medium. Hence a user can access a variety of sound collections by simply installing a different computer readable medium.

Embodiments described herein are directed to a collaborative document environment for reviewing a collection of documents stored in a repository of a document management system. A shared collaborative document area in a virtual world is associated with a corresponding collection of documents in a document management system. The shared collaborative document area is customized based on a semantic context of the documents in the collection of documents.

An executing electronic arrangement for producing foodstuff recommendations and facilitating shopping in grocery stores, includes semantic context management entity configured to maintain semantic context-defining user information relative to a plurality of users, which indicates user preferences relative to groceries based on obtained explicit or implicit preference information, a base item recommendation entity configured to determine, for a first user, a plurality of more generic grocery recommendations based on the semantic context of the first user and the semantic contexts of a number of other users considered as neighbors to the first user according to predetermined similarity criteria, and a product recommendation entity configured to derive, for the first user, a shopping list incorporating a plurality of recommended products relative to a predetermined store offering such products, representing more specific instances of the determined more generic grocery recommendations and being derived utilizing knowledge of the store's product range or current stock.

A system and method for generating a frame of reference for a plurality of information, the plurality of information containing text data and obtained by a user through interaction with one or more information sources. The method and system include receiving selected information for analysis, the information including a plurality of text data and identifying a plurality of logical units of the text data. Also included are identifying a plurality of individual textual portions in each of the logical units and calculating the number of logical units associated with each of the individual textual portions of the plurality of textual portions for use in identifying a plurality of patterns including a respective pattern for each of the individual textual portions. Based on the identified patterns, a measure of importance for each of the plurality of patterns is calculated and respective patterns from the plurality of patterns are selected based on their corresponding measure of importance satisfying a predefined pattern importance threshold for retaining the respective patterns in the plurality of patterns. Further, a plurality of information context definitions is generated based on the selected respective patterns and the generated information context definitions are assigned as context definitions of a semantic context associated with the frame of reference. The plurality of information is related to each other and such relation is represented by the generated context definitions of the semantic context.

Communication of a compressed message over a communication channel between message processors. The compressed message may be expressed in terms of an expressed or implicit template identification, and values of one or more parameters. Based on the template identification, the meaning of the one or more parameters may be understood, whereas the meaning of the parameter(s) may not be understood without a knowledge of the template. The template provides semantic context for the one or more parameters. The transmitting message processor may have compressed the message using the identified template. Alternatively or in addition, the receiving message processor may decompress the message using the identified template. The template itself need not be part of the compressed message as transmitted.

A method of, and system for, generating a sentence from a semantic representation maps the semantic representation to an unordered set of syntactic nodes. Simplified generation grammar rules and statistical goodness measure values from a corresponding analysis grammar are then used to create a tree structure to order the syntactic nodes. The sentence is then generated from the tree structure. The generation grammar is a simplified (context free) version of a corresponding full (context sensitive) analysis grammar. In the generation grammar, conditions on each rule are ignored except those directly related to the semantic representation. The statistical goodness measure values, which are calculated through an analysis training phase in which a corpus of example sentences is processed using the full analysis grammar, are used to guide the generation choice to prefer substructures most commonly found in a particular syntactic / semantic context during analysis.

A method, apparatus and computer program product for presenting a user interface for a conversational system is described. A unified semantic representation of conversational content between a user and conversational system is created as a contextual graph of concepts and relations. A set of subgraph components of the semantic contextual graph dynamically identified based on a current dialog activity. The identified set of subgraph components in a user interface as a set of graphical elements representing respective concepts and relations.

The various embodiments herein provide a method and a system for building an entity hierarchy. The method comprises extracting a plurality of entities from a bin data, determining a parent entity by understanding a context in which the entity is used, resolving the entities by bringing the synonymous entities together and holding the polysemous entities apart based on a semantic context and a syntactic context and building a hierarchical structure of entities using knowledge repositories, ontologies and language repositories along with natural language processing techniques. The method of extracting entities from the structured data comprises identifying each data point as an entity and identifying entities based on a relationship defined with other entities. The method of extracting entities from unstructured data includes a self-learning process and training based learning process to learn new parent entities from domain specific documents using new entity recognition models.

The invention provides a method and a device for intelligently correcting input information. According to the method comprising the steps of collecting a context word segmentation sequence of a training sample; extracting a feature vector of the training sample based on the training sample and a word vector of the context word segmentation sequence of the training sample; acquiring an error detection model according to a feature vector training classifier; and judging whether the input information to be detected comprises wrong word segmentation according to the error detection model, and if yes, correcting the wrong word segmentation, the technical problem that an error correction rate is not high due to the fact that the semantic context of the input information is not considered during existing correction on the input information is solved, intelligent correction on the input information by using the error detection model is achieved, and error correction efficiency and the error correction rate are improved.

The invention relates to a computer-implemented method of training a neural network, comprising: training a first neural network (4) of a self organizing map type with a first set (2) of first text documents (3) each containing one or more keywords (7) in a semantic context to map each document (3) to a point (X i / Y j ) in the self organizing map (5) by semantic clustering; determining, for each keyword (7) occurring in the first set (2), all points (X i / Y j ) in the self organizing map (5) to which first documents (3) containing said keyword (7) are mapped, as a pattern (6) and storing said pattern (6) for said keyword (7) in a pattern dictionary (9); forming at least one sequence (11) of keywords (7) from a second set (12) of second text documents (13) each containing one or more keywords (7) in a semantic context; translating said at least one sequence (11) of keywords (7) into at least one sequence (14) of patterns (6) by using said pattern dictionary (9); and training a second neural network (15) with said at least one sequence (14) of patterns (6). The invention further relates to computer-readable media and classification, prediction and translation machines based on neural networks.

The invention relates to a method and a system for extracting entity relations from texts, a storage medium and an electronic device. The method comprises the following steps: acquiring entity triplerelation set, entity and entity attribute set, and concept set; Training a triple relation set of a sentence of a text set and two entities identified in the sentence; Obtaining a sentence comprisinga training text set, two entities identified in the sentence, concepts corresponding to the two entities and a relation set of the two entities respectively, inputting sentence vectors into an entityrelation extraction model and training; Each sentence contains two entities, concepts corresponding to the two entities, and a set of relationships between the two entities. The entity relationship extraction method of the invention extracts the relationship between entities by using the semantic context information in the text, and solves the error labeling problem existing in the remote monitoring process.

A method and system are provided for assisting a user with Emoji in communication content. The method includes receiving, by a user interface, an input string that includes one or more Emoji characters for interpretation based on a user profile. The method further includes verifying and recovering, by a processor-based Emoji semantic verification and recovery system, an intended meaning of the one or more Emoji characters in the input string, by applying rules based on the user profile and natural language processing to analyze the input string for semantic context, interpret the one or more Emoji characters based on the semantic context, and replace at least one of the one or more Emoji characters with one or more other characters to form a modified input string.

A computer-implemented method of training a neural network includes training a first neural network of a self organizing map type with a first set of first text documents each containing one or more keywords in a semantic context to map each document to a point in the self organizing map y semantic clustering; determining, for each keyword in the first set, all points in the self organizing map to which first documents containing said keyword are mapped, as a pattern and storing said pattern for said keyword in a pattern dictionary; forming at least one sequence of keywords from a second set of second text documents each containing one or more keywords in a semantic context; translating said at least one sequence of keywords into at least one sequence of patterns using the pattern dictionary; and training a second neural network with the at least one sequence of patterns.

The invention relates to a video fast forward / fast backward speed self-adaptive regulating method based on semantic content. The video fast forward / fast backward speed self-adaptive regulating method comprises the following steps: firstly, extracting a shot semantic set of each shot by taking the shot as a unit, then, obtaining semantic context of a video shot in a supervisory manner, and representing the context through a context tag tree; then, regulating weight of the shot by the context tag tree of a video; determining the fast forward / fast backward rate of each shot according to the weight of the shot. The video fast forward / fast backward speed self-adaptive regulating method based on semantic content disclosed by the invention can be used for regulating the video fast forward / fast backward speed in a self-adaptive manner according to the video semantic content and the context of each shot to help a user to quickly skip uninterested content; moreover, the interested content is not missed due to excessively high skip speed, so that convenience is provided for the user.

Clinical Information Systems (CIS) may combine evidence-based and best-practice diagnosis and treatment of care delivery into an electronic workflow with clinical decision support (CDS) tools (e.g. order sets, structure documentation, rule-based alerts, compliance and performance reports) to improve care delivery. Configurating CIS's has required manually intensive processes and clinical experts to determine mapping of clinical concepts in the CDS to the CIS. Changes in medicine change configuration requirements causing changes to data structure, semantics, and even historical information. Traditional mapping is non-explicit and hard coded with a high effort cost for reconfiguration, interoperability, and quality assessment. Using Semantic Context Maps, simultaneously mapping naturally stated, native encoding, and semantic encoding in a clinical context, provides simplification of configuration of CDS and increases value of clinical information as medicine evolves. The invention constructs semantically encoded clinical data using semantic context mapping to enable analytics using semantic processing instead of traditional query analytics.

A method and a system computationally performs scene analysis of semantic traffic spaces based on an adaptive spatio-temporal ray-based approach. The method includes acquiring a spatial semantic environment map including semantic context data, calculating at least one feature for at least one location on the spatial environment map taking into account the semantic context of the location and determining a category for the at least one location based on the at least one calculated feature. A feature is a ray based feature calculated by integrating input values along at least one ray extending in least one of a space dimension and a time dimension. The ray may have a ray shape corresponding to a general shape of a road and / or the at least one spatially extending ray follows a course of the road and / or is perpendicular to a surrounding road segment orientation.

Described herein is a computer implemented method for creating content for electronic dictionaries. An exemplary system includes a user interface, entry filtration system, and interface tools for dictionary entry comparison, entry merge, and visual markup of changes. Many dictionaries may be accessed and used in one user interface window. A user may enter a grammatical, syntactic and semantic markup which may be helpful when the user translates a word or a text directly from an electronic document. An appropriate lexical meaning may be selected during translation from among several lexical meanings depending on a grammatical, syntactic and / or semantic context of a word or phrase.

The invention relates to the field of e-commerce, and discloses a method of intention entity recognition based on a user query log. The method comprises steps: an original log is analyzed and extracted, and a session query set, a common click query set and a common query commodity set are formed; through processing, mutual disjoint subsets (I, Q) are obtained, a bipartite graph is built through a click relation, and word clustering is carried out; according to a word clustering result, segment processing is carried out on each query word, and multiple segments are obtained; and weight calculation is carried out on the segments, weight scores of the segments are further obtained, and a segment with the highest score serves as the corresponding entity of the query word. Through analyzing the original log, the query word is subjected to word clustering and weight analysis, intention entity recognition is realized, a context and a semantic context are blended, the query accuracy is improved, better intention recognition effects are achieved, and the online calculation overhead is saved.

The invention discloses an intelligent question and answer sentence pair semantic matching method and device for government affair consultation service, and belongs to the technical field of artificial intelligence and natural language processing. The technical problem to be solved by the invention is how to capture deeper semantic context features and interaction information between sentences andreduce the loss of semantic information, so intelligent semantic matching of sentence pairs is realized. The adopted technical scheme is as follows: a sentence pair semantic matching model composed of a multi-granularity embedding module, a gating deep feature residual fusion network module, a semantic feature interaction matching module and a label prediction module is constructed and trained; and gated deep feature residual fusion expression of sentence information is achieved, meanwhile, final matching tensors of sentence pairs are generated through an attention mechanism and a gating mechanism, and the matching degree of the sentence pairs is judged, so that the purpose of intelligent semantic matching of the sentence pairs is achieved. The device comprises a sentence pair semantic matching knowledge base construction unit, a training data set generation unit, a sentence pair semantic matching model construction unit and a sentence pair semantic matching model training unit.

The invention relates to a time correlation interactive fusion-based online community post influence sorting method. The method comprises the following steps of: determining time correlation-based influence sorting; carrying out semantic modeling on text contents of posts according to a text semantic model, so as to construct a semantic context similarity-based semantic tree; and fusing text semantic meanings of the posts with structure information of a network community according to a uniform interactive fusion algorithm, and sorting the influences of the posts. Through fusing time correlation information into a sorting scheme, a time correlation influence sorting manner is formed, a uniform fusion framework which combines the semantic meanings and the network community structure is realized on such basis, and the network structure is fused with the text contents, so that the problem of discovering the potential influence sources in the network community is solved, and displayed time information and semantic context information of network community posts can be sufficiently utilized.

The invention provides a human body behavior recognition method based on multilayer depth features. The human body behavior recognition method comprises the following steps of training a deep learningmodel on a target database; inputting the sample into a deep learning model, and extracting a top convolutional layer feature map and top full connection layer features; respectively carrying out maximum pooling operation on each channel of the top convolutional layer feature map, and connecting pooling results into a column vector; and connecting a column vector formed by the pooling result witha top layer full connection layer feature to serve as a final video feature expression, and finishing a behavior identification task by combining a support vector machine. The method is based on characteristics of a top full connection layer and a top convolution layer of a deep learning model. The classification capability of the full connection layer of the deep learning model and the semanticcontext description capability of the convolutional layer are considered, and the complementarity and respective advantages of the multi-layer features are fused, so that the identification capabilityof video feature expression is improved, and the behavior identification precision and the operation efficiency of the algorithm are improved.

A system and method permit analysis and generation to be performed with the same reversible probabilistic model. The model includes a set of factors, including a canonical factor, which is a function of a logical form and a realization thereof, a similarity factor, which is a function of a canonical text string and a surface string, a language model factor, which is a static function of a surface string, a language context factor, which is a dynamic function of a surface string, and a semantic context factor, which is a dynamic function of a logical form. When performing generation, the canonical factor, similarity factor, language model factor, and language context factor are composed to receive as input a logical form and output a surface string, and when performing analysis, the similarity factor, canonical factor, and semantic context factor are composed to take as input a surface string and output a logical form.

The invention discloses a hierarchical human body analysis semantic segmentation method with edge constraint, and mainly solves the problems of poor adaptability and insufficient acquisition of rich human body structure semantic context information in a complex analysis scene in the prior art. The method comprises the following steps: collecting and marking portrait data, and preprocessing training data in the portrait data; and constructing a hierarchical human body analysis semantic segmentation model with edge constraint, training and evaluating the model by using the preprocessed training data, and inputting a picture containing a human body into the trained hierarchical human body analysis semantic segmentation model with edge constraint and meeting the evaluation requirement to obtain a segmentation result picture. According to the method, the edge feature information of the human body analysis semantic segmentation task can be constrained, the phenomena of wrong division and missing division of the human body image by the human body analysis semantic segmentation model in a complex scene are improved, the effect of human body analysis semantic segmentation is effectively improved, and the method can be used for real-time early warning of an intelligent security and protection system and virtual reality reloading of electronic commerce.

A method of, and system for, generating a sentence from a semantic representation maps the semantic representation to an unordered set of syntactic nodes. Simplified generation grammar rules and statistical goodness measure values from a corresponding analysis grammar are then used to create a tree structure to order the syntactic nodes. The sentence is then generated from the tree structure. The generation grammar is a simplified (context free) version of a corresponding full (context sensitive) analysis grammar. In the generation grammar, conditions on each rule are ignored except those directly related to the semantic representation. The statistical goodness measure values, which are calculated through an analysis training phase in which a corpus of example sentences is processed using the full analysis grammar, are used to guide the generation choice to prefer substructures most commonly found in a particular syntactic / semantic context during analysis.