166 results about "Semantic matchmaking" patented technology

The invention discloses a method for constructing a cross-modal knowledge graph, and relates to the technical field of knowledge graphs, and the method comprises the steps: obtaining multi-source heterogeneous data, and carrying out the collection, structural analysis, modal recognition and classification, cleaning and standardization processing, so as to form structured multi-modal data; extracting entities and relationships of the identified modals from the structured multi-modal data, and summarizing the entities and relationships to form a multi-modal knowledge element set; mapping different modal entities in the multi-modal knowledge element set to a unified semantic space, and generating a unified entity relationship set through semantic matching, alignment and fusion; and normalizing the data into knowledge triples, and storing and organizing the knowledge triples through a graph database to form a cross-modal knowledge graph. According to the method, the problems of difficulty in multi-modal heterogeneous information alignment and difficulty in entity relationship extraction can be relieved, semantic association is enhanced, and knowledge graph integrity and multi-scene adaptability are improved.

The embodiment of the invention provides an intelligent factory semantic decision generation method and system based on a knowledge graph, and the method comprises the steps: obtaining an operation data set of an intelligent factory, carrying out the semantic feature extraction of the operation data set, and generating the target semantic feature of an equipment operation parameter and the context correlation feature of a semantic description text; and based on a pre-constructed knowledge graph structure, performing dynamic semantic matching processing on the target semantic features and the context association features, generating a semantic decision instruction set corresponding to the equipment operation parameters, generating an equipment control strategy set according to instruction priorities and instruction execution conditions in the semantic decision instruction set, and sending the equipment control strategy set to a server. And feeding back the equipment control strategy set to a control system of the intelligent factory to trigger operation optimization operation, and updating a semantic node association relationship in the knowledge graph structure. According to the method, the problem of fragmentation of equipment operation state representation is effectively solved, and the interpretability of feature extraction is improved by utilizing a collaborative verification mechanism of numerical parameters and text description.

Techniques for multimodal document retrieval are disclosed herein. Multimodal documents that include both textual and graphical components are retrieved from a knowledge base by a multimodal retrieval augmented generation (RAG) agent in response to a query. The documents and / or components or chunks thereof are retrievable by the RAG agent from the knowledge base using the semantic summaries and / or vector search of embeddings in the knowledge base that are generated from text extracted from processing non-textual components of the data. The RAG agent classifies the query type to determine whether to use a semantic match for text or image summaries, full text semantic search, vector cosine similarity search, and / or other multimodal vector search. The RAG agent performs types of searches selected based on the modality used to generate the response to the query.

The invention relates to the technical field of intelligent recommendation, and discloses an AI intelligent matching method based on a knowledge graph, and the method comprises the steps: constructing a quaternary knowledge graph containing a time dimension; identifying legal entities and semantic relationships by adopting an entity identification and relationship extraction technology; multi-level semantic features are extracted through a two-layer progressive semantic matching algorithm of a grammar layer, a semantic layer and a reasoning layer; constructing lawyer ability portraits based on a heterogeneous graph neural network and a time sequence perception graph convolution technology; progressive matching calculation is adopted, the optimal matching weight is learned through a multi-layer attention mechanism, and dynamically optimized intelligent matching is achieved. The technical problems of cold start, insufficient semantic understanding ability and poor timeliness processing ability in the existing legal consultation matching system can be solved, and the matching precision and the user satisfaction are improved.

The invention relates to a knowledge graph enhanced multi-modal file retrieval method, and belongs to the field of artificial intelligence and multi-modal information retrieval. The invention aims to solve the problems of multi-modal information semantic segmentation, weak semantic reasoning ability and low semantic matching precision in the existing multi-modal archive resource retrieval process. Through four stages of archive multi-modal data preprocessing and feature extraction, knowledge graph construction and enhancement, semantic retrieval request analysis and intention modeling, and multi-modal semantic matching and sorting, a semantic relationship is enhanced by utilizing a knowledge graph, so that the semantic relevancy and context consistency of a retrieval result are remarkably improved; a user is allowed to input and inquire in various forms such as texts, images and voices, and semantic extension retrieval is supported. According to the method, more efficient and accurate archive resource retrieval can be realized, and the user retrieval experience and archive knowledge utilization are improved.

The invention provides an elevator multi-mode crowd feature perception and intelligent advertisement putting method and system, and relates to the technical field of intelligent advertisement putting, and the method comprises the steps: collecting multi-mode perception data in an elevator through an edge computing terminal; performing feature decoupling on the data, performing cross-modal semantic alignment, establishing a directed association relationship between modals, and constructing a scene feature map; calculating the topology importance degree of map nodes, screening feature nodes, and extracting context information for semantic coding; mapping the scene semantic code and the advertisement audience semantic code to a two-dimensional coordinate system to construct a semantic matching graph, and extracting an optimal matching path to form a candidate set; predicting a scene evolution trend based on the scene characteristic spectrum evolution trajectory, and calculating an advertisement adaptive score to generate a playing sequence; putting and collecting user interaction data feedback according to the sequence to update the graph structure. According to the invention, accurate crowd feature recognition and advertisement dynamic matching are realized, and the advertisement putting efficiency and the user experience are improved.

Systems and methods are provided to improve data retrieval from a cache memory by using semantic matching to retrieve data from the cache memory. The system includes a two-tiered cache system, with a first tier implementing “key-value” pairs, and a second tier that includes a table that is configured as an artificial intelligence (AI) search indexed source. When a new input does not have a matching “key” at the first tier, the system performs a semantic search at the second tier of the cache to determine if relevant data is stored in the cache. The current systems and methods increase the likelihood of obtaining data for queries from the cache memory, reduce the response time to the queries, improve search consistency, reduce computing resource utilization, improve system performance, and reduce costs.

The invention discloses a bidding and tendering knowledge graph construction method and system based on OCR and NLP. The method comprises the following steps: collecting a multi-source bidding document, and converting an unstructured document into text data by utilizing OCR (Optical Character Recognition); key entities, relations and attributes are extracted through the NLP technology; constructing a knowledge graph with entities as nodes and relationships as edges, and generating entity node vectors; analyzing the bidding and tendering file, extracting text demands such as qualification requirements and scoring standards, and converting the text demands into demand vectors; performing semantic matching on the demand vector and the entity node vector, and calculating the similarity; and extracting high-similarity entities and attribute relationships thereof from the knowledge graph based on a matching result, and automatically filling and generating bidding document core chapters according to a preset template. The system can realize structured storage and intelligent reuse of bidding and tendering knowledge, and improves the bidding document compiling efficiency and quality.

The invention relates to the technical field of video conferences, and discloses a real-time document collaborative editing and synchronizing method in a video conference, which is used for solving the problem of insufficient concurrent editing conflict solving mechanism in the traditional method. The method comprises the following steps: firstly, receiving document editing operations submitted by a plurality of participants through a terminal, and establishing an operation sequence; comparing the timestamps with the position information, identifying concurrent operation groups as conflict sets, and arranging the conflict sets according to a time sequence; types and contents are extracted from the conflict set, semantic matching is carried out, matching operation is converted into a compatible sequence, and subsequent position offset is adjusted; distributing the compatible sequence to a synchronous buffer area, updating a server document state, and generating an incremental update package; a packet is pushed to the online terminal, application changes are buffered locally, and sequence playback is provided for a new participant; the monitoring terminal confirms the signal, re-deduces the unconfirmed change, and circularly checks the Hash consistency; according to the method, conflict resolution is optimized through semantic matching and offset adjustment, and efficient synchronization is achieved.

The invention discloses an AI-based digital project performance evaluation data processing method and system, and relates to the technical field of artificial intelligence and digital project management, and the method comprises a multi-source data collection module which is used for obtaining data; the entity alignment module is used for establishing an association relationship among cross-system data entities; the dynamic index generation module is used for dynamically adjusting the evaluation index weight by utilizing a reinforcement learning framework; the efficiency prediction module is used for modeling a task dependency relationship according to the time sequence diagram convolutional network and outputting a delay risk probability; and the visual interface is used for displaying the data. According to the method, cross-system entity alignment, dynamic index weight adjustment, task delay risk prediction and root cause analysis are realized through AI technologies such as BERT semantic matching and a graph neural network, visual data display and interaction functions are provided by virtue of a visual interface, the problems of data dispersion, index static state, risk lag and the like in traditional project evaluation are solved, and the project evaluation efficiency is improved. And the decision support capability of project management is improved.

The invention discloses a large model calling system and method based on an MCP tool chain, and the method comprises the steps: receiving user input, and carrying out the preliminary matching of tools: querying whether the user input contains an entity in an entity database or not through a fuzzy matching technology, and carrying out the preliminary screening of a plurality of candidate tools based on a matching result, so as to obtain limited to-be-selected tools; user input is converted into semantic vectors, semantic matching is conducted in a vector database, and semantic similarities of all tools to be selected are ranked through a similarity calculation method; extracting key information in user input; abstracting the user input into a specified format; and matching the specified format with the historical workflow template, and if the historical workflow template is matched, replacing the key information to a corresponding position of the historical workflow template by adopting a structure-preserving parameter replacement algorithm to realize workflow reuse. According to the method and the device, a complete technical link from accurate tool screening to intelligent workflow multiplexing can be realized.

The invention relates to the technical field of industrial communication protocol analysis, in particular to a protocol semantic matching analysis method based on reinforcement learning, and the method comprises the steps: constructing an industrial interaction environment containing industrial control equipment, a communication module and a protocol simulation assembly, defining a state space containing a flow attribute and a communication state, field processing, semantic annotation and dependency configuration related action space are carried out; calculating a reward value through a multi-level reward function of dynamic weight adjustment; based on the preprocessed multi-source protocol data set, adopting a Transform and graph neural network dual model to pre-train and generate a template library, and embedding an intelligent agent state processor; the intelligent agent updates the strategy after state processing, action distribution output by the strategy network, verification and execution of the action executor and interaction with the environment; and storing the interaction experience for iteration, and finally outputting protocol structure analysis, field classification and abnormity prompt information. According to the method, low-manual-dependence automatic analysis, unknown protocol self-adaptive adaptation and high-precision semantic matching can be realized.

A variety of smart inference systems and robotic control devices are disclosed. A device networking system interconnects mobile devices, robotic devices, and others. A robotic controller secures and manipulates a controlled device user interface. A smart device system interprets data based on a plurality of models operating on a plurality of runtimes based on a plurality7 of capabilities. A smart payment processor system assigns payments accounts automatically as a function of a semantic matching between one or more inferred semantic identities. A robotic emulation device captures and analyzes a video signal from a target device and transmits manipulation signals emulating a peripheral input device. A system of carts are physically coupled in a charging configuration. A conveyor is configured to couple with another conveyor and having sensors to detect weight, size, location or other aspects of conveyed items, and implementing semantic analysis for management and augmentation.

The invention discloses an industrial control equipment vulnerability identification method and system based on a double-BERT model, and the method comprises the steps: an asset analysis stage: analyzing the flow to generate structured asset data; in the semantic matching stage, a multilingual Sension-BERT model is used for extracting vectors and matching a vulnerability library; in the version identification stage, the BERT is finely adjusted to judge the influence condition of the version; and in the vulnerability confirmation stage, an identification result is manually rechecked. The system comprises an asset processing subsystem and a vulnerability intelligent identification subsystem, and the asset processing subsystem comprises a flow analysis module, an equipment fingerprint identification module, an asset information management module, a scanning task scheduling module and a vulnerability display and screening module. And the vulnerability intelligent identification subsystem comprises a semantic vector matching module, a version matching judgment module and an identification result output module. According to the method, the problems of high asset isomerism, weak semantic matching capability and low version comparison accuracy in the current industrial network can be solved, and the vulnerability identification quality and efficiency are improved.

The invention provides an intelligent recommendation method and system for scientific and technological achievement transfer, and the system comprises a data processing module which is used for collecting and preprocessing achievements, enterprise demands, and third-party data; the knowledge graph module is used for constructing a knowledge graph through entity recognition and relation extraction and supporting incremental updating; the intelligent recommendation module is used for calculating the semantic association degree, the semantic matching degree and the collaborative filtering degree, generating a comprehensive recommendation score through weighted fusion and carrying out sorting recommendation on the scientific and technological achievements; and the feedback and optimization module is used for collecting user feedback scores to dynamically optimize weight distribution. According to the method, the knowledge graph and multi-index fusion are utilized, recommendation accuracy and adaptability are improved, and scientific and technological achievement transfer is promoted.

The invention discloses an automatic artistic advertisement design method and system based on AI, and relates to the technical field of advertisement design, and the method comprises the steps: firstly, extracting visual features and text features from historical advertisement materials, constructing a creative gene vector, and building a creative feature map based on the visual features and the text features; secondly, a creative popularity parameter is calculated by utilizing user interaction data, and weighted optimization is carried out on the advertisement creativity by using real-time feedback information; secondly, semantic matching is carried out on the target product features and the creative gene vectors, and a transportability score is calculated through corrected semantic relevancy and historical validity data; and finally, the most suitable creative gene is selected, content fusion and rendering are carried out, and final advertisement content is generated. Based on dynamic optimization of creativity popularity and user feedback, the real-time performance and accuracy of the advertisement effect are improved, the cold start problem of new product advertisement creativity is effectively solved, and cross-platform and cross-domain creativity migration and adaptation can be achieved.

The invention relates to the technical field of computers, discloses a production command case-oriented interactive deduction method and system, and aims to solve the problems of static solidification of deduction cases, low degree of freedom of interaction, disjunction with real-time production data and low case construction efficiency in the prior art. The method comprises the following steps: constructing a digital twinborn model synchronized with a physical environment in real time; receiving an event injection or parameter adjustment instruction input by a user; driving a multi-agent simulation engine to simulate production process evolution based on the instruction, and generating process simulation data; analyzing the simulation data and quantifying the influence of the decision on the production indexes to form a performance evaluation report; searching a semantic matched historical case knowledge graph, and extracting a disposal plan; according to the scheme, scientific deduction and intelligent assistance of production command decisions are realized, emergency response efficiency is improved, precipitation and reuse of command experience are promoted, multi-user collaborative deduction is supported, and team decision collaborative ability is enhanced.

Techniques for multimodal document retrieval are disclosed herein. Multimodal documents that include both textual and graphical components are retrieved from a knowledge base by a multimodal retrieval augmented generation (RAG) agent in response to a query. The documents and / or components or chunks thereof are retrievable by the RAG agent from the knowledge base using the semantic summaries and / or vector search of embeddings in the knowledge base that are generated from text extracted from processing non-textual components of the data. The RAG agent classifies the query type to determine whether to use a semantic match for text or image summaries, full text semantic search, vector cosine similarity search, and / or other multimodal vector search. The RAG agent performs types of searches selected based on the modality used to generate the response to the query.

The invention provides a semantic search method and system for blood disease medical record big data, and relates to the technical field of medical information processing.The method comprises the steps that firstly, a blood disease medical record semantic association network is constructed, medical record text units serve as nodes, semantic association relationships serve as edges, and related information of the nodes and the edges is marked; analyzing a user search intention to generate a search semantic vector, calling semantic association network information to perform dynamic semantic matching to obtain a matched medical record text unit set, generating a hierarchical search result according to a matching result, dividing and sorting according to a matching dimension, and generating semantic association interpretation; and finally, based on the access behavior feedback of the user to the hierarchical search result, optimizing the semantic association network, recording user clicking, staying and secondary search triggering conditions, and adjusting the node association priority, edge association strength and weight to improve the accuracy and practicability of blood disease medical record search.

The invention discloses a multi-source data management evaluation method based on soil and agricultural products, and relates to the technical field of agricultural data management, and the method comprises the steps: collecting preprocessing data, constructing semantic network association, calibrating a sensor, extracting features, constructing a causal network, and carrying out the comprehensive evaluation and updating of a model, and achieves the management evaluation. The method has the advantages that the self-adaptive semantic network containing the soil, agricultural product and environment domain term ontology library is constructed, the semantic mapping relation is updated in real time based on the ontology evolution algorithm, the dynamic semantic matching degree is calculated in combination with the context sensing analysis engine so as to achieve multi-source data dynamic association alignment, edge end calibration nodes are deployed, and the data dynamic association alignment accuracy is improved. Real-time dynamic calibration of sensor data is completed through an environmental factor compensation model and a drift correction formula, the problem of inaccurate evaluation caused by insufficient dynamic association of multi-source data is systematically solved, and finally accurate management and scientific evaluation of soil and agricultural product multi-source data are realized.

The invention provides a man-machine conversation-based software cost intelligent evaluation interaction method and system, and relates to the technical field of software cost evaluation, and the method comprises the steps: obtaining project description information, carrying out intention recognition and entity extraction, and carrying out semantic matching in combination with a knowledge graph to obtain candidate cost references; according to the method, the estimation reference value is generated based on multi-dimensional similarity calculation and weight self-adaptive adjustment, user feedback correction is supported, accurate evaluation and dynamic adjustment of the software cost are achieved, and the accuracy, transparency and efficiency of software cost evaluation are improved.

The invention provides a method for joint answering of a user knowledge base and a service system under an AI platform, and belongs to the technical field of AI platform models.According to the method, hierarchical semantic matching is achieved by constructing a three-level knowledge vector cache and a three-level question intention vector cache and adopting vector representation of different dimensions; after user query is received, a semantic similarity algorithm is used for performing parallel retrieval to generate an initial matching degree, multi-dimensional similarity information is fused through a joint matching degree calculation function, and the precision is improved by using an intention recognition optimization model; calculating an optimal sequence length through an adaptive parameter adjustment function according to the long and short memory matching length to obtain real-time data of the service system, fusing the static answer of the knowledge base and the real-time data of the service system to generate a final answer, and updating a vector cache and optimizing algorithm parameters according to user feedback. And high-precision joint answering of the knowledge base and the business system is completed.

The invention discloses an automatic form filling method, which comprises the following steps of: by defining a target field set, carrying out standardized preprocessing on to-be-filled information of a user and related materials to obtain a user standardized information key value pair; a target webpage is accessed, in the access process, a DOM tree structure and a complete visual image of the target webpage are obtained, the DOM tree structure and the complete visual image of the target webpage are input into the multi-modal large model to be analyzed, and a multi-modal feature vector is obtained; based on the multi-modal feature vector, identifying interactive elements in the target webpage to obtain an interactive element set; semantic matching is carried out through attributes of interactive elements and unique keys of user standardized information key value pairs, a mapping relation is established, and an interactive element set belonging to a target field set is screened out; and defining an automatic processing step sequence set, traversing the automatic processing step sequence set, and executing an automatic filling operation.

The invention relates to the technical field of knowledge maps, in particular to a personalized question and answer recommendation method based on a knowledge map and two-layer attention, and the method comprises the steps: constructing and storing a structured knowledge network containing a plurality of entities and association relationships thereof, analyzing an input natural language query, and recognizing query elements; based on semantic comprehension, calculating a semantic matching degree between the query elements and candidate answers extracted from the structured knowledge network, and further identifying a deep semantic intention of the natural language query; and based on the identified query elements and semantic intentions, in combination with the semantic matching degree, performing retrieval in the structured knowledge network to obtain an initial result, performing structured recombination and natural language generation on the initial result, and outputting a readable answer. The method has higher question and answer matching precision and query generation quality in a data sparse and relation complex scene, and is widely applied to the fields of education question and answer, intelligent recommendation, knowledge retrieval and the like.

Automatic post-editing of machine translated content is disclosed herein. An example method includes receiving a source document for translation, calculating semantic signatures of text chunks within the source document using an artificial intelligence (AI) model, matching the semantic signatures against a repository of previous translations, and displaying a document-based visualization that includes both semantic and exact matches found in the repository of previous translations.

The invention relates to the field of engineering consultation service quality control, and discloses a service quality evaluation system, which comprises a data acquisition module, a dynamic topology construction module, a dual-channel evaluation module, an improvement suggestion generation module and an evaluation report generation module. A dynamic topology network with engineering documents as nodes and multi-dimensional association as edges is constructed, deep semantic matching of a real-time compliance detection channel and quality state differential equation prediction of a deep evolution analysis channel are combined, and standard compliance quantitative evaluation and quality evolution trend analysis are achieved. The abnormal propagation path detection based on the topology gradient field is linked with the industry knowledge graph to generate optimization suggestions, and the multi-source evaluation data is integrated into a standardized report through a dynamic association mapping technology. According to the method, the structured, dynamic and intelligent levels of engineering consultation service quality evaluation are improved, and the method has good practicability and engineering adaptation capability.

The invention discloses a task book demand standardization method based on retrieval enhancement generation, and the method comprises the following steps: S1, gathering historical demand documents and standardized records, completing cleaning, term unification and structured storage, and forming a maintainable and searchable domain knowledge base; s2, segmenting the document into entries, carrying out keyword labeling and vectorization representation and indexing, and establishing a complete knowledge base system; s3, generating a high-quality reference set through pre-retrieval enhancement, semantic matching retrieval and post-retrieval enhancement; s4, based on the multi-layer template, driving the large model to output a structured demand confirmation document draft; and S5, providing interactive auxiliary revision containing user feedback and a double termination mechanism, and providing powerful intelligent support for finally forming a demand document. According to the method, the defects of element recognition, knowledge reuse and consistency guarantee in the prior art can be overcome, and a feasible engineering path is provided for large-scale and cross-scene demand standardization.

The invention provides an adaptive tracking method for a complex scene. The adaptive tracking method comprises the following steps: respectively extracting template features and search features; similarity matching features are obtained through calculation, and feature fusion is carried out on the similarity matching features and the search features to obtain self-adaptive fusion features; respectively inputting the adaptive fusion features and the search features into a preset candidate region generation network to select a plurality of candidate regions; performing refined semantic matching calculation on the template features and each candidate region to obtain similarity features; respectively obtaining a matching semantic score and a coordinate frame of the similarity between each candidate region and the template in the matching semantic branch, and obtaining a foreground probability score and a coordinate frame of each candidate region as a foreground target in the foreground probability branch; calculating to obtain a punishment score of each candidate region; and performing weighted calculation on the penalty score, the matching semantic score and the foreground probability score of the same candidate region to obtain a final score of each candidate region, and selecting the candidate region with the highest score as a tracking target.

The invention is suitable for the technical field of computers, and provides an integrated code generation method which comprises the steps that on a low-code platform, a seed code is generated based on a data model, and the seed code comprises service metadata capable of being recognized by a development environment; analyzing a data model and a business rule on which the seed code is based, and constructing a multi-dimensional meta-knowledge graph; in a code editor, analyzing the business metadata to generate a business context; deducing a service constraint condition from the service context, and converting the service constraint condition into a code generation constraint; on the basis of the code generation constraints, code snippets are generated through an AI programming assistant in combination with the seed codes, and therefore semantic matching is carried out on data relations, service rules and the like in services and constraint conditions needed by code generation through integration of low-platform codes, a code editor and the AI programming assistant, deep semantic understanding is achieved, and the service quality is improved. And generating a corresponding code snippet.

Systems and methods are provided to improve data retrieval from a cache memory by using semantic matching to retrieve data from the cache memory. The system includes a two-tiered cache system, with a first tier implementing "key-value" pairs, and a second tier that includes a table that is configured as an artificial intelligence (AI) search indexed source. When a new input does not have a matching "key" at the first tier, the system performs a semantic search at the second tier of the cache to determine if relevant data is stored in the cache. The current systems and methods increase the likelihood of obtaining data for queries from the cache memory, reduce the response time to the queries, improve search consistency, reduce computing resource utilization, improve system performance, and reduce costs.