32 results about "Semantic framework" patented technology

A speech dialogue service apparatus including: a language analysis module tagging a part of speech (POS) of each respective word included in a sentence recorded in a predetermined text, syntactically analyzing the sentence by classifying a meaning of each respective word, and generating at least one semantic frame corresponding to the sentence according to a result of the syntactical analysis; and a dialogue management module analyzing an intention of the sentence corresponding to the at least one respective semantic frame, and generating a system response corresponding to the sentence intention by selecting a predetermined sentence intention according to whether an action corresponding to the intention of the respective sentence can be performed.

A method of probabilistic error-tolerant natural language understanding. The process of language understanding is divided into a concept parse and a concept sequence comparison steps. The concept parse uses a parse driven by a concept grammar to construct a concept parse forest set by parsing results of speech recognition. The concept sequence comparison uses an error-tolerant interpreter to compare the hypothetical concept sequences included by the concept parse forest set and the exemplary concept sequences included in the database of the system. A most possible concept sequence is found and converted into a semantic framed that expresses the intention of the user. The whole process is led by a probability oriented scoring function. When error occurs in the speech recognition and a correct concept sequence cannot be formed, the position of the error is determined and the error is recovered according to the scoring function to reduce the negative effect.

A speech dialogue service apparatus including: a language analysis module tagging a part of speech (POS) of each respective word included in a sentence recorded in a predetermined text, syntactically analyzing the sentence by classifying a meaning of each respective word, and generating at least one semantic frame corresponding to the sentence according to a result of the syntactical analysis; and a dialogue management module analyzing an intention of the sentence corresponding to the at least one respective semantic frame, and generating a system response corresponding to the sentence intention by selecting a predetermined sentence intention according to whether an action corresponding to the intention of the respective sentence can be performed.

The invention relates to a method for automatically recognizing the semanteme of natural language sentences understood by a computer, in particular to a method for accurately recognizing the Chinese language, which comprises the following specific steps: a, establishing a body base in a certain filed; b, establishing a semantic framework knowledge base on the basis of a field body; and c, directly matching the natural language sentences with semantic structures on the basis of the body mapping of semantic frameworks, and recognizing the matching according to the modes of the semantic frameworks. The method is very different from the mainstream method of segmenting words in the second-generation search engine technology. The segmented words have concept annotations of the field body, the accurate semantic matching of the natural language sentences can be obtained hereby, and the computer system can carry out calculation and inference on the body knowledge, therefore, the deep artificial intelligent question and answer has a wide prospect of application.

The invention provides a speech and semantic recognition method of a service robot. The method comprises: system word library modeling is carried out; to be specific, a word library database is established and then a semantic frame database is established; a lip language mode base is established; a system uses a speech recognition system to collect a speech instruction inputted by a user and a face video and speech and semantic recognition and video segment lip language recognition are carried out respectively; and according to a combined recognition result, a determined Chinese semantic information is displayed at a display interface. Therefore, a noun database and a semantic frame database are established for different service fields; and on the basis of the word library database, the semantic frame database, and the lip language mode base, the robot understands the Chinese meaning expressed by the speech instruction, so that speech and semantic recognition becomes accurate.

The invention discloses a system and a method for indexing and retrieving keywords of academic documents. The system is based on an academic thesis network submission editing system and an academic document database retrieval system. The keywords at the inlet of the network submission editing system construct an input interface according to semantic frames which qualitatively describe the completesubject content of the documents such as research objects, conditions, methods, processes, attributes, results and the like. An author of a thesis directly indexes journal papers, and a word list isautomatically generated by a computer so as to realize synonymy, homoionym and cross-language automatic indexing integration and retrieving integration of the keywords. The retrieval system establishes an optional retrieval inlet according the same keyword essential data items and sequence thereof, the position relation of the words determines the definition relation of the words so as to improvespecificity.

A framework for generating a semantic interpretation of natural language input includes an interpreter, a first set of types, and a second set of types. The interpreter is adapted to mediate between a client application and one or more analysis engines to produce interpretations of the natural language input that are valid for the client application. The first set of types is adapted to define interactions between the interpreter and the one or more analysis engines. The second set of types is adapted to define interactions between the interpreter and the client application.

The invention provides a BIM and GIS integration method based on semantic fusion. The BIM and GIS integration method comprises the steps: S1, building a unified semantic framework of BIM and GIS integration, and forming a shared domain ontology; S2, obtaining instance information from the BIM data model to fill the shared domain ontology; S3, semantizing the GIS data and filling the GIS data intoa shared domain ontology in an ontology alignment mode; and S4, realizing information query and reasoning based on the fused domain knowledge model. The BIM and GIS integration method researches a method for converting IFC and CityGML into an RDF graph model, so as to enable the heterogeneous data to be expressed by a semantic-based knowledge model, thus solving the fusion problem of heterogeneousinformation of BIM and GIS, classifies, maps and hierarchically organizes classes and relations in the field to form concept integration under a unified framework, well processes and displays representation of knowledge, and infers implicit knowledge through establishment of logic rules.

The invention discloses a method for forwarding data of a hierarchal wireless sensor network based on semantic routing, wherein the hierarchal wireless sensor network is a semantic network. The method comprises the steps of (1) a cluster member node collecting data in the data collection period; (2) the member node processing the collected data, comparing the collected data with a preset threshold, uploading the collected data to a cluster head node if the collected data is larger than the threshold, otherwise, discarding the data and waiting for the next data collection period to collect data; (3) the member node determining the semantic type according to the collected data, constructing related semantic frame of the data, and directly forwarding the constructed data packet to the cluster head node; (4) the cluster head node extracting data according to the data packet uploaded by the cluster member node in one data collection period to fuse the data, determining whether an event occurs according to the data fusion result, and determining whether or not to upload and forward the data packet of the related event; and (5) the cluster head node determining the semantic type, constructing an event semantic frame, writing semantic elements contained in the occurred event of the cluster into the event semantic frame, and the cluster head node carrying out data packet construction according to the generated event semantic frame, and directly injecting the constructed data packet to the network through the routing table.

The invention relates to a semantic frame-based power grid defect text mining method. Characteristic information in a power grid defect text is mined for meeting various defect automatic statistic demands. An existing statistic method is based on labor, so that the time and labor are wasted and the correct rate is subjectively influenced to a great extent. The method comprises the steps of firstly extracting syntactic structure knowledge of the defect text and constructing an ontology dictionary; secondly filling semantic slots with keywords of the defect text by applying a semantic slot filling method; thirdly integrating the disordered slots into a semantic frame by utilizing a semantic association algorithm; and finally performing word string combination to realize simplification of the semantic frame. The defect automatic statistics of different demands can be conveniently realized after the semantic frame is constructed. The method is high in defect text statistic accuracy and convenient to apply.

The invention discloses a structured memory graph network model for multiple rounds of spoken language understanding, which is composed of an input coding layer, a memory coding layer, a feature aggregation layer and an output classification layer, dialogue behaviors generated by spoken language understanding tasks are used to replace texts as memory nodes for coding, and the dialogue behaviors are formatted representations containing semantic framework information. And the unstructured characters are converted into a structured triple. A graph attention network is used for replacing a recurrent neural network and an attention mechanism to achieve feature aggregation, sequence information between the attention mechanism and dialogue nodes is reserved, and model learning how to effectivelyutilize structured memory nodes is facilitated. According to the network model, the encoding dialogue behavior replaces a historical dialogue text to serve as a memory unit, original information of asemantic framework is reserved to the maximum extent, and the problems that in the prior art, noise is generated in complex occasions due to the fact that text information depends on a model, and operation efficiency is low are solved.

The invention relates to a hierarchical machine translation method and system based on predicate argument structures. The hierarchical machine translation method comprises training and translating processes, and comprises the following steps: organically organizing all predicate argument structures in sentences into a graphic structure; taking the graphic structure as a top semantic framework structure for improving hierarchical phrase machine translation to directly build the models of the predicate argument structures in the hierarchical phrase machine translation, separately translate fragments with independent semantics and combine the fragments into a final translated text according to structures among the fragments. The translated text obtained by using the method is better in semantic similarity, semantic structure and long-distance ordering characteristics, thereby reducing the probability of generating meaningless translated texts through machine translation.

The invention discloses a man-machine conversation method and system based on a semantic framework. The method comprises a step of creating a subject forest structure tree according to an original corpus and generating a semantic frame model by using the subject forest structure tree, and mapping an entity attribute of the subject forest tree to a corresponding semantic slot in the semantic framemodel, a step of matching a subject type for a visitor question in man-machine conversation and filling the semantic slot in the semantic frame model corresponding to the subject type with the visitorquestion, a step of mapping the visitor question of the filled semantic slot in the semantic frame model to the entity attribute of the subject forest structure tree, and a step of allowing the subject forest structure tree to carry out question matching in a knowledge base according to the visitor question, and feeding an answer corresponding to the matched question back to a visitor. Thus, an accurate and complete visitor question can be obtained, on the above basis, the accuracy of the answer is ensured and the communication efficiency is improved, the system can actively interact with thevisitor, and the user experience is improved.

Disclosed are a question answering system for structured knowledgebase using deep natural language question analysis, and a method thereof, the question answering system for structured knowledgebase using deep natural language question analysis includes a deep natural language question analysis unit configured to create a structure of a semantic frame by analyzing a natural language question that is input, a question-intermediate expression creation unit configured to create a question-intermediate expression of a lexicon level based on the semantic frame, a knowledgebase-specialized query creation unit configured to create a query used to search in knowledgebase that is a subject of search, based on the question-intermediate expression, and a knowledgebase search unit configured to find a correct answer in the knowledgebase that is subject of search based on the query, to provide an accuracy of the correct answer, a confidence of the correct answer and an evidence for the correct answer.

The embodiment of the invention discloses an intelligent law enforcement lawyer assistant customer service system. The intelligent law enforcement lawyer assistant customer service system comprises anatural language processing module used for converting natural language characters input by a user into a semantic framework model which can be understood by the system, and a user intention understanding module used for receiving the semantic framework output by the natural language processing module and processing the semantic framework according to a preset method so as to identify the intention of a user. The above module also comprises a question and answer module including a single-round retrieval sub-module and a case result estimation sub-module; and an answer generation module used for forming structured answer information from answers of questions and answer module and returning the structured answer information to the user. The intelligent law enforcement lawyer assistant customer service system is flexible in interaction mode, can realize multiple rounds of conversations, and can provide case cases with reference values according to conditions.

A user input processing apparatus for a vehicle includes: an operation command generator recognizing a first semantic frame from a user voice, generating a first operation command based on the recognized first semantic frame, recognizing a second semantic frame from the user voice, and generating a second operation command based on the recognized second semantic frame; a sequence determination unit generating an execution sequence of the first operation command and the second operation command; and a controller controlling an execution of the first operation command and the second operation command based on the generated execution sequence and a relation between the first operation command and the second operation command.

A task generation method and apparatus for corpus processing are disclosed. The method includes establishing a semantic framework according to a domain; Establishing a corpus script on the basis of the semantic framework; Selecting a seed template according to the corpus script; And determining a seed template needed for the reformulation and generating a reformulation corpus crowdsourcing task. The present application solves the technical problem that the task generation method for corpus processing has poor effect. Crowd-sourced task are generated and high-quality corpus are obtained throughthat present application. In addition, it is possible to complete the task of rewriting crowdsourcing which is suitable for the collection of Chinese rewriting corpus at a lower cost by means of theapplet. Thus, user feedback can be obtained more dynamically and crowdsourcing quality can be improved.

The invention relates to a digital standard meta-model based on semanteme and an application method thereof, the meta-model comprises three parts of data contents, namely DSS general ontology, technical description data and structure and data constraint, which are established based on a unified semantic framework, and semantic elements in the unified semantic framework comprise DSS composition classes, DSS object relationships, DSS object attributes and DSS object instances; the DSS composition class is divided into a primary class and a secondary class and is used for concluding concepts. According to the method, content expression of standard elements is achieved based on unified semantics, DSS is automatically read, understood and processed through a computer, automatic comparison between DSS models or DSS examples is achieved, conformity of products or processes and batch products is automatically judged, and reasoning decision and automatic control are conducted based on the DSS; through registration, reference and inheritance mechanisms, sharing and reusing of standard elements are achieved, and the efficiency of digital design and standard system revision and the standard implementation benefits are improved.

There is provided a system and method for processing and / or recognizing acoustic signals. The method comprises obtaining at least one pre-existing speech recognition model; adapting and / or training the at least one pre-existing speech recognition model incrementally when new, previously unseen, user-specific data is received, the data comprising input acoustic signals and / or user action demonstrations and / or semantic information about a meaning of the acoustic signals, wherein the at least one model is incrementally updated by associating new input acoustic signals with input semantic frames to enable recognition of changed input acoustic signals. The method further comprises adapting to a user's vocabulary over time by learning new words and / or removing words no longer being used by the user, generating a semantic frame from an input acoustic signal according to the at least one model, and mapping the semantic frame to a predetermined action.

The invention provides a machine-aided dialogue system, and the system comprises a semantic framework device which is configured to provide one or more dialogue themes; a knowledge base memory which is configured to store semantic knowledge information, wherein the semantic knowledge information comprises at least one group of characters, words, phrases, sentences, paragraphs or combinations of the characters, the words, the phrases, the sentences and the paragraphs; a dialogue management device which is configured to host a dialogue process according to a dialogue theme provided by the semantic framework device, wherein the process hosting dialogue comprises the step of generating one or more question expression statements of any topic in the one or more dialogue topics based on semanticknowledge information obtained from the knowledge base memory, and determining that the current dialogue theme is completed in response to a preset dialogue ending condition, and ending the current dialogue theme.

The invention discloses an interactive knowledge learning method based on a semantic framework. The method comprises the following steps: step 1, adding a curiosity model; step 2, maintaining a credibility value for each channel and the user according to the ability and credibility of question answering and different knowledge fields, and accumulating confidence coefficients of different channels for the generated knowledge; 3, redesigning a semantic network storage structure, and dividing a knowledge storage space according to confidence; 4, constructing a statistical cognition module; 5, feeding back results to different channels and users according to the screening condition of the knowledge, and adjusting the values of the credibility of the corresponding fields to form a dynamic knowledge construction closed loop; and step 6, reflecting the credibility of each channel or user on the confidence coefficient of knowledge in the next knowledge construction process. According to the method, storage of knowledge with different levels of confidence degrees is divided and isolated, and the knowledge is dynamically constructed in an external interaction process, so that the obtained formal knowledge is more effective.

The invention relates to the technical field of semantic recognition, discloses an implementation method of a general semantic framework, and solves the problems of server resource waste and high development time cost and labor cost of the existing semantic recognition framework. The method comprises the following steps of: when a framework is started for the first time, obtaining configuration components of each skill from Redis of a database, checking whether configuration components conform to a general configuration specification, generating configuration information of corresponding skills for the configuration components conforming to the general configuration specification, and loading the configuration information to Metadata of the framework; calling a component initialization method to initialize the configuration components, and loading the configuration components into a memory after the initialization is completed; and developing an access interface, an addition skill interface, a deletion skill interface and an upgrade skill interface for the framework. The method not only reduces the waste of server resources, but also reduces the time cost and labor cost of development, and is suitable for semantic recognition.

The invention discloses a construction method of a cloud computing system based on super resource integration. Construction behaviors of the cloud computing system are uniformed to be two kinds of unification: super resource uniforming of system components, super resource integration uniforming of construction operation and semantic framework of integration. The super resource uniforming of the system components refers to uniforming the components of the cloud computing system into super resources; the super resource integration uniforming of construction operation refers to uniformly viewing forming and running behaviors of the cloud computing system as integration operation, and the cloud computing system is formed by combined, compositing and integrated integration operation; and the semantic framework of integration refers to enabling the integration operation to form multistage layered general characteristic components by criss-cross layered abstraction and prefabricating a sharable standard component for forming specific integration operation. The method has the advantages of formal semantics of cloud operation and crisscrossing of the process, and development efficiency and product quality of the cloud computing system can be remarkably improved simultaneously.

The invention provides a knowledge verification method and system based on an information system. The method comprises the steps that relations between knowledge view element model concepts are determined by building a knowledge view element model and determining a semantic frame and concepts of the knowledge view element model, an extended knowledge view model is obtained, and then a knowledge representation system is built according to the extended knowledge view model; whether all relation nodes x in the knowledge representation system has legal relations or not is judged, and then the integrity and correctness of the knowledge representation system is verified. The purposes that one information system can represent existing knowledge in a standard, complete and logical mode, verify the integrity and correctness of the knowledge representation system, and increase the knowledge reuse rate are achieved.

The invention discloses a human-computer dialogue method and system based on a semantic framework, which creates a theme forest structure tree based on original corpus and uses the theme forest structure tree to generate a semantic framework model, and maps entity attributes of the theme forest structure tree to the semantic framework The corresponding semantic slot in the model; during the man-machine dialogue, match the topic type of the visitor's question, and fill the visitor's question into the semantic slot in the semantic framework model corresponding to the topic type; The visitor's question in the slot is mapped to the entity attribute of the theme forest structure tree; finally, the theme forest structure tree performs question matching from the knowledge base according to the visitor's question, and feeds back the answer corresponding to the matched question to the visitor; thus ensuring accurate On this basis, it can ensure the accuracy of answers and improve communication efficiency, and can actively interact with visitors to increase user experience.

The invention relates to the technical field of intelligent language processing, in particular to a construction method of a PCB fault diagnosis rule base, electronic equipment and a storage medium thereof. The PCB fault diagnosis rule base construction method comprises the steps of obtaining a plurality of fault statements of an execution object, performing grammatical analysis on the plurality of fault statements to determine a verb type, filling a semantic framework with a grammatical analysis result according to the verb type, performing semantic analysis on information in the semantic framework to obtain risk information, and storing the risk information in a semantic template form, and converting the information in the semantic template into a target language to obtain a fault aiming rule base. According to the PCB fault diagnosis rule base construction method and the established rule base provided by the invention, the diagnosis of the fault of the PCB based on the natural language can be realized, the PCB information in the IPC standard can be applied to the fault diagnosis to a greater extent, and an important reference significance is provided for the construction of an expert system knowledge base in the future.

The invention relates to a hierarchical machine translation method and system based on predicate argument structures. The hierarchical machine translation method comprises training and translating processes, and comprises the following steps: organically organizing all predicate argument structures in sentences into a graphic structure; taking the graphic structure as a top semantic framework structure for improving hierarchical phrase machine translation to directly build the models of the predicate argument structures in the hierarchical phrase machine translation, separately translate fragments with independent semantics and combine the fragments into a final translated text according to structures among the fragments. The translated text obtained by using the method is better in semantic similarity, semantic structure and long-distance ordering characteristics, thereby reducing the probability of generating meaningless translated texts through machine translation.

The invention provides a semantic framework construction method based on human-computer interaction. The method is characterized by comprising the following steps of: 1, realizing the construction of a semantic framework by using a CFN framework; 2, making implementation based on the theory of framework semantics; step 3, referring to work of an English framework network of Berkley; 4, extracting evidence support from the Chinese large corpus; and 5, constructing a framework semantic relationship on the basis. According to the system and the method, the feeling of onsite interaction of teachers and students is improved, the content introduction of works is richer, knowledge points are relatively more concentrated, the information integration degree is also improved through man-machine interaction, the situation that the students spend a large amount of time and energy to search related contents is avoided, and the teaching efficiency is improved.

The invention relates to a semantic frame-based power grid defect text mining method. Characteristic information in a power grid defect text is mined for meeting various defect automatic statistic demands. An existing statistic method is based on labor, so that the time and labor are wasted and the correct rate is subjectively influenced to a great extent. The method comprises the steps of firstly extracting syntactic structure knowledge of the defect text and constructing an ontology dictionary; secondly filling semantic slots with keywords of the defect text by applying a semantic slot filling method; thirdly integrating the disordered slots into a semantic frame by utilizing a semantic association algorithm; and finally performing word string combination to realize simplification of the semantic frame. The defect automatic statistics of different demands can be conveniently realized after the semantic frame is constructed. The method is high in defect text statistic accuracy and convenient to apply.

The invention relates to the technical field of blockchain finance, in particular to an efficient digital asset transaction method based on an account model. Aiming at an existing AFCoin technology framework, a blockchain forest and endorsement node selection method is introduced, a receipt processing process is integrated, complete implementation steps of the framework are given, and the framework is applied to the field of digital asset transaction, so an efficient digital asset transaction method based on an account model is obtained. Under a semantic framework of AFCoin, the blockchain forest is a plurality of blockchain structures maintained by a central bank, and different commercial banks have different blockchains, but are published by the central bank in a unified manner; the endorsement node selection method ensures that a commercial bank cannot select a specific commercial bank responsible for endorsement, and prevents cheating between commercial banks. When a central bank is used as an issuer of digital assets, a commercial bank is used as a seller of the digital assets, and a user is used as a holder of the digital assets, an efficient digital asset transaction methodcan be obtained.