31 results about "Functional semantics" patented technology

The invention discloses a material object programming method and a material object programming system, which belong to the field of human-machine interaction. The method comprises the following steps of: 1) establishing a set of material object programming display environment; 2) shooting the sequence of material object programming blocks which are placed by a user and uploading the shot image toa material object programming processing module by using an image acquisition unit; 3) converting the sequence of the material object blocks into a corresponding functional semantic sequence by usingthe material object programming processing module according to the computer vision identification modes and the position information of the material object programming blocks; 4) determining whether the current functional semantic sequence meets the grammatical and semantic rules of the material object display environment or not, and if the current functional semantic sequence does not meet the grammatical and semantic rules of the material object display environment, feeding back a corresponding error prompt; 5) replacing the corresponding material object programming blocks by using the useraccording to the prompt information; and 6) repeating the steps 2) to 5) until the functional semantic sequence corresponding to the sequence of the placed material object programming blocks meets the grammatical and semantic rules of the material object display environment, and finishing a programming task. By using the method and the system, the problem that children and green hands are difficult to learn programming is solved, and the system has low cost and is easy to popularize.

The invention discloses an entity programming method and system based on an infrared and wireless transmission technology and belongs to the field of man-machine interaction. The entity programming method comprises the following steps: 1) placing entity programming blocks by a user according to a task and a task execution rule of the entity programming; 2) sending a binary identification code of one entity programming block to the adjacent entity programming block through an infrared module; 3) polling a wireless module of each entity programming block by a wireless processing module, thereby obtaining entity programming block sequence information and sending the entity programming block sequence information to an entity programming processing module; 4) converting binary sequence information into a corresponding functional semantic sequence by the entity programming processing module and judging if the present functional semantic sequence meets a grammar semantic rule of the present entity display environment; and 5) finishing the task of the entity programming when the functional semantic sequences corresponding to all the entity programming blocks meet the grammar semantic judging rule of the entity programming display environment. The entity programming method and system provided by the invention are easily realized; the error feedback is avoided; and the efficiency is high.

Extraction of functional semantics and isolated dataflow from imperative object oriented languages, in one aspect, may include identifying one or more methods and / or classes associated with one or more of a plurality of property labels in a computer code written in object oriented language to extract functional and isolation characteristics in the computer code. The plurality of property labels supported by one or more checking rules, are used to verify that the one or more methods and / or classes identified with the plurality of property labels have isolation characteristics. An object oriented language compiler is provided for supplying the plurality of property labels and checking rules. The object oriented language compiler further may include capability to transform methods into compute tasks and connect the compute tasks so as to create a dataflow graph.

Disclosed are a method and a system for voice recognition based on environmental parameter and group trend data. The method includes the steps of firstly, acquiring user's voice information and environmental parameter and identification information relative to the voice information and identifying the user according to the identification information; secondly, extracting personal voice and behavior habit characteristic data which are stored in a personal voice and behavior habit characteristic libratory while extracting group voice and habit characteristic data, of a similar group, stored in a group voice and habit characteristic libratory, and judging which functional semantic analysis modules are adaptable to the voice information and sending the relative information to the semantic analysis modules to do individual or combined semantic analysis to obtain a semantic analysis result; and thirdly, storing the semantic analysis result and sending the semantic analysis result and the relative information to the personal voice and behavior habit characteristic libratory and the group voice and habit characteristic libratory to classify and file. By the method and the system, accuracy of voice recognition is improved greatly.

The invention discloses a code abstract generation method based on a code knowledge graph and knowledge migration. The method comprises the following steps: 1, extracting and fusing multi-source datato construct the code knowledge graph; 2, generating an abstract syntax tree by using a code parsing tool, and traversing method calling nodes in the abstract syntax tree to obtain an API calling sequence; 3, searching API description information in the API calling sequence from the code knowledge graph; 4, mining large-scale open source projects, constructing an API sequence abstract generation data set, training a text abstract generation model, and learning background knowledge related to API functions; 5, constructing a code abstract generation model; and 6, training a code abstract generation model on the code abstract data set, and generating a code abstract by utilizing the model. According to the method, the code knowledge graph and knowledge migration are used for the code abstract generation task, the ability of the model to extract code functional semantic information can be enhanced, the model is assisted in understanding codes, and the code abstract generation quality is improved.

The invention relates to the technical field of web services and aims to improve availability of semantic web service descriptive models and achieve semantization of existing WSDL-described web services. According to the technical scheme, the method includes the steps of 1, acquiring functional semantics from a WSDL-described service document; 2, acquiring context semantics from the WSDL-described service document; 3, calling a web service, defining performance semantics of the web service according to MSDL standards; 4, acquiring time-space semantics of the service from the WSDL-described document; 5, acquiring technological semantics of the service from the WSDL-described document; 6, defining interaction semantics of the web service according to MSDL standards. The performance semantics of the multi-dimensional model describes quality attribute of the service. The method is mainly applied to web service technologies.

The invention relates to the technical field of web services and aims to provide a web service multi-dimensional semantic model, achieve semantization of web services, and increase availability of a web service semantic descriptive model. According to the technical scheme, a model building method includes: 1, defining functional semantics of web services according to MSDL standards; 2, defining context semantics of web services according to the MSDL standards; 3, defining performance semantics of web services according to the MSDL standards and describing quality attributes of the services; 4, defining technical semantics of web services according to the MSDL standards; 5, defining interaction semantics of web services according to the MSDL standards. The method is mainly applied to web services.

The invention discloses ontology-based product function semantic modeling representation, which comprises the following steps: (1) the product function modeled by OWL-DL language defining the ontologyconcept of product functional semantics, and obtaining the ontology of functional semantics; (2) redefining the functional verbs in the functional base as ontological concepts, and using OWL-DL ontology language adds semantic constraint to realize the semantic expansion of the concept of functional verbs; (3) creating SysML metamodel consistent with functional semantic ontology; (4) using the SysML metamodel created in step (3) to create the function definition model and the function structure model. The ontology-based product function semantic modeling representation provided by the invention can eliminate the uncertainty and ambiguity of the function representation, provide a semantic structure understandable by a computer, and lay a foundation for applications such as function reasoning and automatic function decomposition.

Extraction of functional semantics and isolated dataflow from imperative object oriented languages, in one aspect, may include identifying one or more methods and / or classes associated with one or more of a plurality of property labels in a computer code written in object oriented language to extract functional and isolation characteristics in the computer code. The plurality of property labels supported by one or more checking rules, are used to verify that the one or more methods and / or classes identified with the plurality of property labels have isolation characteristics. An object oriented language compiler is provided for supplying the plurality of property labels and checking rules. The object oriented language compiler further may include capability to transform methods into compute tasks and connect the compute tasks so as to create a dataflow graph.

Described is a system for understanding machine-learning decisions. In an unsupervised learning phase, the system extracts, from input data, concepts represented by a machine-learning (ML) model in an unsupervised manner by clustering patterns of activity of latent variables of the concepts, where the latent variables are hidden variables of the ML model. The extracted concepts are organized into a concept network by learning functional semantics among the extracted concepts. In an operational phase, a subnetwork of the concept network is generated. Nodes of the subnetwork are displayed as a set of visual images that are annotated by weights and labels, and the ML model per the weights and labels.

The invention discloses a building three-dimensional model semantization method based on a software drawing. The building three-dimensional model semantization method comprises the steps that S1, a known geographic area coordinate system is established; s2, building function semantic classification; s3, constructing a textualized semantic model; s4, building roof semantic recognition and verification of a textualized semantic model; and S5, judging unknown building types and establishing an unknown city database. According to the method, semantic recognition of the building can be accurately carried out on the city level, so that urban water, electricity and gas and the building are guided, and intelligent management and application of search purposes of other industries are assisted.

The invention discloses a service semantic relationship mining method, which comprises the following steps: mining parameter relationship; merging specific services with identical functional semantics to generate abstract services, wherein the identical functional semantics refers to identical semantics of input parameter sets and output parameter sets which form the specific service operation; mining the relationship between abstract services based on comparison between the input parameter sets and the output parameter sets; verifying whether specific service relationship exists, if yes, persisting into a database. The method can efficiently organize and manage the web services, can efficiently narrow the searching space during service discovery and service composition, brings convenience for realization of automatic service discovery and composition, and meanwhile, provides information for a service provider for maintaining vital services and developing new services by means of analytical investigation on service networks.

The invention discloses a material form change functional decomposition method based on flow semantic. The method includes the steps that 1, a meta model and a material form type model library are established; 2, level form graphs are adopted to represent the input form and the output form of materials based on the meta model and a material form type model library; 3, level form graphs of the input form and the output form are compared, and local changes corresponding to the overall form changes of the materials are obtained; 4, the local changes are subjected to priority ranking; 5, local change operations are executed in sequence according to the priority, and the input form is converted into the output form. In the initial stage of concept design, functional semantic including material flow information is subjected to modeling representation, and functional decomposition is conducted through an established model, and the change process of the materials is deduced.

The invention discloses a material object programming method and a material object programming system, which belong to the field of human-machine interaction. The method comprises the following steps of: 1) establishing a set of material object programming display environment; 2) shooting the sequence of material object programming blocks which are placed by a user and uploading the shot image toa material object programming processing module by using an image acquisition unit; 3) converting the sequence of the material object blocks into a corresponding functional semantic sequence by usingthe material object programming processing module according to the computer vision identification modes and the position information of the material object programming blocks; 4) determining whether the current functional semantic sequence meets the grammatical and semantic rules of the material object display environment or not, and if the current functional semantic sequence does not meet the grammatical and semantic rules of the material object display environment, feeding back a corresponding error prompt; 5) replacing the corresponding material object programming blocks by using the useraccording to the prompt information; and 6) repeating the steps 2) to 5) until the functional semantic sequence corresponding to the sequence of the placed material object programming blocks meets the grammatical and semantic rules of the material object display environment, and finishing a programming task. By using the method and the system, the problem that children and green hands are difficult to learn programming is solved, and the system has low cost and is easy to popularize.

The invention discloses a function-level defect positioning method based on an embedding technology, and the method comprises the steps: employing a code embedding technology based on an abstract syntax tree to represent function code function semantics, and employing a word embedding technology to represent defect report problem semantics; finally, fusing the function and the semantic features ofthe defect report by using a convolutional neural network, and predicting a suspicious function related to a given defect. In order to solve the problem of limited training data, a pre-training modelis proposed to perform feature representation on defect reports and codes; meanwhile, aiming at the problem that the number of class instances is unbalanced, a random oversampling method is providedfor processing the class instances. Tests on three mainstream Java projects find that when the recommendation list is 10, the accuracy of the method can reach 12.5%-40%, and the method has great potential in the field of fine-grained defect positioning and has great potential application value on mainstream Java software projects.

The invention discloses a Mashup service feature representation method based on functional semantic association calculation. The method comprises the following steps: 1, preprocessing all Mashup service data needing feature representation; 2, based on the preprocessed Mashup service data, performing a function noun extraction operation; 3, for the function noun set FS of each Mashup service, carrying out semantic association calculation on the semantic weight of each function noun; and 4, based on the semantic weight calculation result in the step 3, combining a TF-IDF algorithm and a Word2Vecmodel to express a Mashup semantic feature vector. According to the invention, the matching precision of Mashup services and the service search efficiency can be effectively improved.

The invention discloses a tower data processing method and device based on a functional semantic coordinate system. The method comprises the steps of acquiring a first position parameter value of a to-be-processed tower, wherein the first position parameter value is expressed based on a functional semantic coordinate system, and at least one coordinate axis of the functional semantic coordinate system is used for representing the change direction of a set function of the tower; converting the first position parameter value from the functional semantic coordinate system to a second position parameter value in a position coordinate system of a tower; and according to the second position parameter value, performing set operation processing on the to-be-processed tower. By utilizing the method, the first position parameter value of the tower to be processed is converted into the second position parameter value convenient to process through the functional semantic coordinate system and the position coordinate system, so that the tower to be processed is subjected to setting operation processing, an inspection route does not need to be preset, and the labor cost is reduced.

Described is a system for understanding machine-learning decisions. In an unsupervised learning phase, the system extracts, from input data, concepts represented by a machine-learning (ML) model in an unsupervised manner by clustering patterns of activity of latent variables of the concepts, where the latent variables are hidden variables of the ML model. The extracted concepts are organized into a concept network by learning functional semantics among the extracted concepts. In an operational phase, a subnetwork of the concept network is generated. Nodes of the subnetwork are displayed as a set of visual images that are annotated by weights and labels, and the ML model per the weights and labels.

The invention discloses a material form change functional decomposition method based on flow semantic. The method includes the steps that 1, a meta model and a material form type model library are established; 2, level form graphs are adopted to represent the input form and the output form of materials based on the meta model and a material form type model library; 3, level form graphs of the input form and the output form are compared, and local changes corresponding to the overall form changes of the materials are obtained; 4, the local changes are subjected to priority ranking; 5, local change operations are executed in sequence according to the priority, and the input form is converted into the output form. In the initial stage of concept design, functional semantic including material flow information is subjected to modeling representation, and functional decomposition is conducted through an established model, and the change process of the materials is deduced.

The invention relates to the technical field of web services and aims to improve availability of semantic web service descriptive models and achieve semantization of existing WSDL-described web services. According to the technical scheme, the method includes the steps of 1, acquiring functional semantics from a WSDL-described service document; 2, acquiring context semantics from the WSDL-described service document; 3, calling a web service, defining performance semantics of the web service according to MSDL standards; 4, acquiring time-space semantics of the service from the WSDL-described document; 5, acquiring technological semantics of the service from the WSDL-described document; 6, defining interaction semantics of the web service according to MSDL standards. The performance semantics of the multi-dimensional model describes quality attribute of the service. The method is mainly applied to web service technologies.

The invention discloses a service semantic relationship mining method, which comprises the following steps: mining parameter relationship; merging specific services with identical functional semantics to generate abstract services, wherein the identical functional semantics refers to identical semantics of input parameter sets and output parameter sets which form the specific service operation; mining the relationship between abstract services based on comparison between the input parameter sets and the output parameter sets; verifying whether specific service relationship exists, if yes, persisting into a database. The method can efficiently organize and manage the web services, can efficiently narrow the searching space during service discovery and service composition, brings convenience for realization of automatic service discovery and composition, and meanwhile, provides information for a service provider for maintaining vital services and developing new services by means of analytical investigation on service networks.

A Mashup service feature representation method based on functional semantic association calculation, said method comprising the following steps: the first step, preprocessing all mashup service data that needs feature representation; second step, based on the preprocessed mashup service data , carry out the functional noun extraction operation; the third step, for the functional noun set FS of each mashup service, carry out semantic association calculation on the semantic weight of each functional noun; the fourth step, based on the semantic weight calculation result of the third step, combine TF‑IDF algorithm and Word2Vec model are used to represent Mashup semantic feature vectors. The invention can effectively improve the matching precision of the Mashup service and the service search efficiency.

The invention relates to the technical field of web services and aims to provide a web service multi-dimensional semantic model, achieve semantization of web services, and increase availability of a web service semantic descriptive model. According to the technical scheme, a model building method includes: 1, defining functional semantics of web services according to MSDL standards; 2, defining context semantics of web services according to the MSDL standards; 3, defining performance semantics of web services according to the MSDL standards and describing quality attributes of the services; 4, defining technical semantics of web services according to the MSDL standards; 5, defining interaction semantics of web services according to the MSDL standards. The method is mainly applied to web services.

The invention discloses an entity programming method and system based on an infrared and wireless transmission technology and belongs to the field of man-machine interaction. The entity programming method comprises the following steps: 1) placing entity programming blocks by a user according to a task and a task execution rule of the entity programming; 2) sending a binary identification code of one entity programming block to the adjacent entity programming block through an infrared module; 3) polling a wireless module of each entity programming block by a wireless processing module, thereby obtaining entity programming block sequence information and sending the entity programming block sequence information to an entity programming processing module; 4) converting binary sequence information into a corresponding functional semantic sequence by the entity programming processing module and judging if the present functional semantic sequence meets a grammar semantic rule of the present entity display environment; and 5) finishing the task of the entity programming when the functional semantic sequences corresponding to all the entity programming blocks meet the grammar semantic judging rule of the entity programming display environment. The entity programming method and system provided by the invention are easily realized; the error feedback is avoided; and the efficiency is high.

Disclosed are a method and a system for voice recognition based on environmental parameter and group trend data. The method includes the steps of firstly, acquiring user's voice information and environmental parameter and identification information relative to the voice information and identifying the user according to the identification information; secondly, extracting personal voice and behavior habit characteristic data which are stored in a personal voice and behavior habit characteristic libratory while extracting group voice and habit characteristic data, of a similar group, stored in a group voice and habit characteristic libratory, and judging which functional semantic analysis modules are adaptable to the voice information and sending the relative information to the semantic analysis modules to do individual or combined semantic analysis to obtain a semantic analysis result; and thirdly, storing the semantic analysis result and sending the semantic analysis result and the relative information to the personal voice and behavior habit characteristic libratory and the group voice and habit characteristic libratory to classify and file. By the method and the system, accuracy of voice recognition is improved greatly.

The invention discloses a function-level defect location method based on an embedding technology. The method uses an abstract syntax tree-based code embedding technology to represent function code function semantics, and uses word embedding technology to represent defect report problem semantics. Finally, a convolutional neural network is used to fuse the semantic features of functions and defect reports and predict suspicious functions associated with a given defect. In order to solve the problem of limited training data, the present invention proposes to use a pre-training model to perform feature representation on defect reports and codes; at the same time, it proposes to use random oversampling to deal with the problem of unbalanced number of class instances. Through experiments on three mainstream Java projects, it is found that when the recommended list is 10, the accuracy rate of the method proposed by the present invention can reach 12.5%-40%, showing great potential in the field of fine-grained defect location. It has great potential application value in software projects.

The invention discloses an offline semantic parsing method and system. The method comprises the steps of: obtaining a plurality of functional semantic fields corresponding to a plurality of language fields according to the plurality of language fields, and building a plurality of semantic classification fields skill _ id corresponding to the plurality of functional semantic fields in gram grammarcompiled contents of a language model; judging whether the current field has a plurality of corresponding semantic classification fields skill _ id or not, and if yes, acquiring the corresponding current semantic classification fields skill _ id according to the current field and semantic correspondence; and if not, acquiring a navigation information return field according to the current field anda local navigation analysis program. According to the method, skill _ id fields are added in a gram grammar compiling process, the classification of each statement is indicated, and while an identification result is given, the category of information, namely skill id, to which the current statement belongs is given. The semanteme of the corresponding category is directly positioned through skill_ id for analysis.