30 results about "Semantic matchmaking" patented technology

A method, apparatus and computer program product for creating a dialog system for web content is described. Knowledge is extracted from a target web application for the dialog system. The knowledge includes an organizational structure of the target web application and domain knowledge pertinent to the target web application. A deep learning process associates the domain knowledge with the organization structure of the target application. A plurality of knowledge sources of different respective types are created from the domain knowledge and the organizational structure. Each of the knowledge sources is used for providing answers to user queries to the dialog system. As part of the invention, a semantic matcher is provided to select among the answers provided by the plurality of knowledge sources for a best answer to a user query.

The invention provides a method for publishing advertisement and an advertisement server, the method comprises: performing semantic matching for the advertisement keywords and the media website content to obtain text keywords same as or associated with the advertisement keywords; marking the text keywords in the text of the media website and posting the advertisement content and the link web site corresponding to the advertisement keywords. According to the invention, the media website content can be analyzed, and the advertisement keywords and the corresponding web page text contents concerned by the browsing people are matched accurately by the semantic matching, so that the text keywords same as or associated with the advertisement keywords can be obtained, the advertisement information can be posted on the text keywords, the accuracy for delivering the advertisement by the advertisement user can be effectively improved.

A method, apparatus and computer program product for presenting a user interface for a conversational system is described. A user input is received in a dialog between a user and the conversational system, the user input in a natural language. A domain trained semantic matcher is used to determine a set of entities and a user intent from the user input. One or more queries is generated to selected ones of a plurality of knowledge sources, the knowledge sources created from domain specific knowledge. The results from the one or more queries are ranked based on domain specific knowledge. A system response is presented in the dialog based on at least a highest ranked result from the plurality of knowledge sources.

The invention discloses an application search method and a server. The method comprises the steps of constructing an application tag database, wherein the application tag database comprises tag systems of all applications; constructing a search word tag database, wherein the search word tag database comprises tag systems of a plurality of search words; constructing a tag knowledge base; receiving a current search word uploaded by a client, and obtaining the tag system of the current search word according to the search word tag database; based on the tag knowledge base, calculating semantic correlation between the tag system of the current search word and the tag system of each application; and when the semantic correlation between the tag system of the current search word and the tag system of one application meets a preset condition, returning related information of the application to the client for performing display. According to the scheme, the tag systems of the applications are matched with the tag systems of the search words based on the tag knowledge base, namely, user intentions and the applications are mapped to a same semantic space, so that the problem in semantic matching of the search words and the applications is solved and an application search function is effectively realized.

The disclosed embodiments provide a system for processing user-generated input. During operation, the system obtains a first embedding produced by an embedding model from an input string representing an entity and a hierarchy of clusters of embeddings generated by the embedding model from a set of standardized entities. Next, the system searches the hierarchy of clusters for a subset of the embeddings that are within a threshold proximity to the first embedding in a vector space. The system then calculates embedding match scores between the input string and a first subset of the standardized entities represented by the subset of the embeddings based on distances between the subset of the embeddings and the first embedding in the vector space. Finally, the system modifies, based on the embedding match scores, content outputted in response to the input string within a user interface of an online system.

The invention discloses an interface file processing method and apparatus and a control message processing method and apparatus. The interface file processing method is used for a gateway device. The interface file processing method comprises: performing semantic matching on each device function in a functional body description file of each machine device and each control of a control body instance file; establishing a mapping relationship between the successfully matched device function and control; and for each machine device, according to the mapping relationship, generating a basic file as a man-machine interaction interface for controlling each device function of the machine device, and sending the basic file to a user terminal. According to the interface file processing method, the device function and the control of the machine device are subjected to semantic matching, so that dynamic mapping between the device function and the control is realized; and an interface designer is not required to design an interactive interface, and the design of the interactive interface is finished by cooperation of a gateway and the user terminal, so that the workload of the interface designer is reduced and the interface development efficiency is improved.

Embodiments of the present disclosure disclose a method and an apparatus for storing a media file and for searching a media file. A specific embodiment of the method includes: acquiring a semantic vector for characterizing semantics of a context of the media file, the context being a context of the media file in a webpage presenting the media file; and storing the semantic vector and the media file in association. Based on the corresponding relationship established by this embodiment, the semantic vector corresponding to the media file may be used to match the media file to ensure the semanticmatching of the media file.

The invention belongs to the technical field of network data transmission, and particularly relates to an action recognition method based on a match video. The method comprises the following steps: preprocessing and screening event videos; carrying out keyword labeling and statistics; carrying out action and semantic feature extraction; and performing association training learning on the action and semantic features. According to the action recognition method based on the match video, the match rebroadcasting cost can be reduced, the requirements for personnel and equipment are reduced, a basic model structure based on contents such as machine language automatic recognition and semantic matching is provided for contents such as match rebroadcasting live broadcast, and a technical basis isprovided for realizing competition video-based action recognition for competition rebroadcasting or live broadcasting.

The invention discloses an insurance customer list screening method and device based on health medical big data, belongs to the technical field of semantic processing and semantic analysis of medicalbig data, and aims to solve the technical problem that clients cannot be accurately screened in multiple dimensions due to that fact of single index analysis based on a single data source during userscreening analysis in the prior art. According to the adopted technical scheme, the method specifically comprises the following steps: inputting insurance customer information; searching and extracting medical data, namely searching the medical data of a client from the medical health big data through the client information, and extracting the medical data of the client; performing semantic standardization processing, namely performing semantic standardization processing on the found client medical data to generate client standardized data with a semantic relationship, thereby forming a clientmedical semantic network; performing semantic matching, namely performing semantic matching on the screening rule base and the client medical semantic network; and obtaining client screening results.

The invention relates to the technical field of oil and gas pipeline repair, and discloses a knowledge graph construction method and system for oil and gas pipeline defect repair, which comprises thesteps of collecting related data of pipeline defect repair, performing arrangement and classification according to a pipeline defect repair process to form a preliminary pipeline defect repair data model framework, analyzing and disassembling the pipeline defect repair data model framework, determining a corresponding defect repair business data domain, determining a triple of pipeline defect repair knowledge, identifying the triad, completing knowledge extraction of pipeline defect repair, and importing the identified triad into a graph database to construct a knowledge graph. According to the method, existing pipeline repair implicit knowledge is explicit, pipeline repair business knowledge reference is provided for a user based on semantic search and semantic matching by means of visualgraph data and knowledge reasoning, a foundation is laid for accumulation, transmission and sharing of defect repair knowledge points, and the knowledge service is provided for the user in the defectrepair business activity, and meanwhile, the way of improving the user skills is expanded.

The invention discloses a semantic-based product configuration demand response system and an order generation method thereof. The system comprises a data layer, a semantic layer, a service layer and auser layer, the data layer transmits the product component data and the configuration ontology data to the semantic layer; the semantic layer transmits the semantically processed data to the servicelayer; the service layer transmits the data to the application layer, wherein the data layer comprises a product component database, a configuration ontology database, an in-production information database and an order database; the semantic layer comprises a demand ontology construction module, a semantic matching module, a semantic reasoning query module and a product configuration and order generation module; the service layer comprises a product customization module, an order management module, an order state query module and a product component tracing module, the user layer comprises a client module and an administrator module, the customer requirements are converted into the specific index parameters of the product components to enable customized products to meet the personalized requirements of customers, and the product configuration efficiency is improved.

The invention relates to the technical field of artificial intelligence customer service and intelligent voice analysis, and particularly discloses a text customer service interaction method for the power industry based on NLP, and the method specifically comprises the following steps: step 1, preprocessing voice data; step 2, performing voice recognition, and naturally generating a question-answering system; step 3, generating a semantic database; step 4, semantic matching; step 5, generating characters by a natural language processing method; step 6, outputting the characters .Intelligent voice text interaction is adopted, so that the communication efficiency and experience of customers can be improved, the labor cost is reduced, and high research value is achieved.

The invention discloses a knowledge graph construction method for massive conference texts. The method specifically comprises the following three steps: 1) conference summary extraction: firstly extracting common attributes of each summary document based on a summary template, and then extracting entities and relationships in each summary document based on joint modeling; 2) conference content extraction: firstly extracting common attributes of each conference content document based on a conference content template, then segmenting the non-structured document, and taking each segment as an entity in the knowledge graph; and (3) relation extraction between the conference summary and the conference content: by establishing the relation between the conference summary and the conference content, establishing a relation with each piece of conference content by each conference summary, and extracting the relation by adopting a model based on semantic matching in the process. According to the method, model training is completed based on a self-built data set, a knowledge graph based on conference text data can be constructed, and retrieval of conference information can be completed based on the knowledge graph.

The embodiment of the invention provides a semantic resource updating method. The method comprises the steps that a recognition engine recognizes a dialogue statement input by a user to obtain a recognized text; the intervention engine performs intervention rule matching and intervention semantic matching on the recognition text, and converts the recognition text into recognizable semantic structured rule verbal technique; the semantic engine forwards the rule verbal skill to the associated skill service; and the associated skill service processes the regular verbal skill and feeds back a processing result to the user. The embodiment of the invention further provides a semantic resource updating system. According to the embodiment of the invention, the semantic resource can be subjected toheat treatment, the updating efficiency of the semantic resource is improved, and when the semantic resource has an error, timely repair is facilitated.

The invention discloses a data query method. The method comprises the steps of receiving a to-be-answered question sent by a user terminal; extracting a feature vector of the to-be-answered question by utilizing a pre-established deep semantic matching model to obtain a to-be-matched feature vector; respectively calculating the similarity between the to-be-matched feature vector and each library feature vector; wherein each library feature vector is obtained by performing feature extraction on each service problem group in a target server knowledge base corresponding to the user terminal and performing feature extraction on each general problem group in a general knowledge base in communication connection with each server knowledge base; determining the library feature vector correspondingto the maximum value in the similarities as a target feature vector, and obtaining target answer data of a question group corresponding to the target feature vector; and returning the target answer data to the user terminal. According to the invention, the user experience is improved, the procurement and maintenance costs are greatly reduced, and resources are saved. The invention furthermore discloses a data query device and equipment, and a storage medium, which have corresponding technical effects.

The invention relates to the technical field of computers, and provides a vertical search method and device, electronic equipment and a storage medium which are used for improving the model training effect and further improving the vertical search relevance. The method comprises: after a vertical search request is received, storing a target search keyword contained in the vertical search request in a database; inputting into a semantic matching model obtained based on extended data set training to obtain a search result of a set search type, and then returning a search result page containing the search result. Therefore, the initial training data is expanded through the entity association relationship contained in the service knowledge graph, and the data diversity is increased, so that the model training effect is improved, and the correlation between the search result and the search keyword is improved.