61 results about "Fuzzy query" patented technology

The invention relates to the crossing field of an artificial intelligence technology and an automotive electronic technology, and discloses a voice interaction question answering method and system based on a vehicle-mounted large model knowledge base, and the method comprises the following steps: receiving a user query text, and executing intention recognition and structured analysis; based on an analysis result, constructing and executing a mixed retrieval request including structured filtering, full-text keyword retrieval and semantic vector retrieval; fusing and reordering the plurality of retrieval result sets obtained respectively to obtain a high-quality final retrieval result set; and finally, constructing a context based on the final retrieval result set and a user query text, and calling a large language model to generate and output a final answer. By introducing a hybrid retrieval and fusion reordering mechanism, comprehensive context information is provided for a large language model, the accuracy and reliability of questions and answers are improved, and the ability of a system to understand and process complex fuzzy queries is enhanced.

The invention relates to the technical field of natural language processing, in particular to a verifiable large model retrieval enhancement generation system and method based on an evidence chain, and the method comprises the steps: receiving an initial query, recognizing the fuzziness and information gap of the initial query in combination with an associated retrieval document, and generating a supplementary query set; based on the initial query, the supplementary query and the corresponding retrieval document, generating candidate answers with references and verifying the information supportability of the candidate answers; for the candidate answers passing the verification, extracting support information and constructing a hierarchical attribution mapping relation; and integrating the information to form to-be-evaluated information, and if the current verified to-be-evaluated information meets a preset sufficiency condition, integrating the generated preliminary answer and the to-be-evaluated information to synthesize a target answer. The method effectively overcomes the defects that a traditional RAG system is fragmented in information integration, has one-sided fuzzy query and answer, is low in attribution efficiency and excessively depends on retrieval content, and has the advantages of answer comprehensiveness, verifiability and deployment lightweighting.

The invention relates to the technical field of government affair services, in particular to an NLP entity recognition method based on semantic extension, which comprises the following steps: collecting multi-source time sequence data and generating an entrepreneurship ecological data graph through a dynamic time warping algorithm; dynamic evolution of a term library is realized by using a generative adversarial network, and a term semantic association graph is constructed in combination with a graph convolutional network; a double-path attention mechanism is deployed in a Transform encoder, and global semantics and local attention guided by terms are fused through a gating loop unit; modeling fuzzy query analysis into a Markov decision process, and optimizing a semantic extension strategy by adopting a reinforcement learning agent; multi-granularity features are extracted through a feature pyramid network, and model parameters are optimized in combination with a bidirectional long-short-term memory network and an online learning feedback loop; and finally, calculating entity service association strength by using a graph attention network, and generating an interpretable report based on a Shapley value attribution algorithm. A closed-loop learning system is formed, and the business starting policy matching accuracy is improved.

The invention provides a satellite image fuzzy query method and system based on multi-dimensional vectorization indexing, and relates to the technical field of satellite image retrieval, the method comprises the following steps: mapping a structured query condition generated based on natural language query into a three-dimensional hierarchical index condition structured database to retrieve and output three-dimensional index metadata; and after the three-dimensional index result set is loaded according to the three-dimensional index metadata, space-time interaction visual output for generating the three-dimensional index result set is carried out by taking dynamic weight distribution as a sorting basis. The technical problems that in the prior art, satellite image high-response query cannot be carried out based on user fuzzy expression, so that the satellite image data query and calling efficiency is low, and the query result practicability is insufficient are solved. The technical effects that full-link closed-loop retrieval driven by natural languages is achieved, query flexibility is improved through intelligent mapping, efficient positioning is guaranteed through multi-dimensional indexing, the accuracy of dynamic sorting optimization results is guaranteed, and interaction friendliness is visually enhanced are achieved.

The invention provides an entity enhancement and context-aware paragraph retrieval method for an RAG system for solving the problems of fuzzy query intention and insufficient paragraph context modeling in RAG retrieval. The method comprises the following steps: firstly, identifying and extracting a key entity by using a named entity, carrying out weighted fusion on the key entity and question representation obtained by a pre-training model to form an enhanced query vector, and accurately describing a core semantic intention of the question; secondly, performing semantic modeling on paragraphs in a document library, mining a potential semantic association relationship between the paragraphs, and constructing a context interaction model between the paragraphs based on a graph neural network and a gating loop unit mechanism, so as to obtain paragraph vector representation with complete semantics and clear hierarchy; and finally, calculating the similarity between the enhanced query and the paragraph vector, and completing high-precision paragraph-level retrieval. According to the method, the correlation and the recall rate are remarkably improved, insufficient entity utilization and weak context modeling are relieved, and good expansibility and cross-domain applicability are achieved.

This invention belongs to the field of natural language processing and information extraction technology, and provides a method and system for extracting parameters from boiler design documents based on multi-round self-feedback RAG. The method includes: preprocessing the boiler design document and the original query, converting them into several text segments and expected inputs respectively; encoding the text segments and expected inputs into high-dimensional vectors and query vectors respectively, storing the high-dimensional vectors and constructing a vector index; generating a response based on retrieved external knowledge to assist a large language model, and iteratively optimizing the generated results through multiple rounds. This invention effectively solves the problems of incorrect and missing parameter extraction in complex or fuzzy query scenarios, significantly improving the reliability of the final results; it overcomes the limitations of traditional methods and standard large language models in handling multimodal data, enhancing the processing capability for complex multimodal documents; and it improves the accuracy of information retrieval, laying a solid foundation for subsequent accurate parameter extraction.

The application relates to the technical field of natural language processing, in particular to an evidence chain-based verifiable large model retrieval enhancement generation system and method, which comprises the following steps: receiving an initial query, identifying the ambiguity and information gap of the initial query in combination with associated retrieval documents, and generating a supplementary query set; generating a cited candidate answer based on the initial query, the supplementary query and the corresponding retrieval document, and verifying the information supportability of the cited candidate answer; extracting supportable information from the verified candidate answer and constructing a hierarchical attribution mapping relationship; integrating the above information to form to-be-evaluated information; if the to-be-evaluated information currently verified satisfies a preset sufficiency condition, integrating a generated preliminary answer and the to-be-evaluated information to synthesize a target answer. The application effectively solves the defects of the traditional RAG system, such as information integration fragmentation, one-sided query answering, inefficient attribution and excessive dependence on retrieval content, and has the advantages of comprehensive answer, verifiability and lightweight deployment.

The invention provides an anti-fraud early warning system based on multi-source cross-platform fusion, belongs to the field of network security and privacy computing, and aims to solve the problems of privacy set intersection matching failure and early warning missing report caused by data noise. According to the scheme, data are cleaned through a multi-modal normalization engine, and the cleaning distortion degree is calculated; generating a fuzzy feature vector containing a plurality of variants through a fuzzy feature mapper to avoid the Hash avalanche effect; performing privacy set intersection in a ciphertext domain through an encryption calculation unit; and finally, a risk decision-making module combines the variant weight and the time decay factor to carry out weighted scoring and generate an early warning. According to the method, accurate matching is upgraded into fuzzy query, so that the early warning robustness and accuracy are remarkably improved, and risk quantification and privacy security are realized at the same time.

This invention discloses a method and system for intelligent chart querying based on a large model of multimodal data. The method includes acquiring fuzzy query information input by the user and real-time monitoring data of a machine tool; performing intent recognition and entity extraction on the fuzzy query information, parsing it into structured query conditions; establishing a multimodal data association library based on a large model, inputting the real-time monitoring data of the machine tool into the multimodal data association library, and storing it according to preset association requirements; inputting the structured query conditions into the established multimodal data association library; and generating interactive charts based on the output information of the multimodal data association library, embedding key conclusion summaries. This application utilizes a large model to understand user intent and dynamically recommend related charts, breaking through the traditional fixed menu navigation mode and realizing intelligent interaction through joint analysis of data and charts.

The present application provides a data verification method, device and computer readable storage medium, comprising: obtaining a received key encrypted to-be-verified user identifier, and constructing a fuzzy query condition based on the encrypted to-be-verified user identifier to query the serial number corresponding to the user identifier from a receiver; receiving the query result fed back by the receiver, wherein the query result comprises a plurality of receiver user identifier ciphertext values and corresponding serial numbers matched with the fuzzy query condition; matching the encrypted to-be-verified user identifier with the receiver user identifier ciphertext values in the query result, and if the matching is successful, obtaining the user attribute encrypted by a homomorphic encryption algorithm corresponding to the to-be-verified user identifier based on the corresponding serial number; calculating the difference between the homomorphic encryption value of the user attribute of the requester and the homomorphic encryption value of the user attribute obtained from the receiver through ciphertext calculation, and sending the difference to the receiver; receiving the verification result fed back by the receiver after decrypting and analyzing the difference.

This application provides a method, apparatus, electronic device, and storage medium for generating network topology maps. The method includes: acquiring XDR data, core network configuration data, and B-domain data corresponding to a target enterprise; performing data filtering processing on the XDR data, core network configuration data, and B-domain data to generate a network data field table; constructing a network device service path table corresponding to the network devices within the target enterprise based on the network data field table; and drawing a network topology map corresponding to the target enterprise based on the core network node configuration conversion table and the network device service path table. This application can realize real-time updated full-path visualization applications, fuzzy queries by slice, enterprise name, or DNN, presentation of key indicators and alarms, and achieve functions such as visibility, manageability, and evaluation of private network services, adapting to the needs of refined management of private network resources.

An exemplary system and method are disclosed that provide the private querying of a biometric scan, such as a person's photo, against a private database such that the client issuing the private query learns only the identity of the query if it is in the database while the server performing the processing learns nothing of the biometric scan. The exemplary system and method conduct privacy-preserving searches over fuzzy databases via a fuzzy labeled set intersection (FLPSI) protocol. FLPSI protocol computes an intersection of noisy input sets to assess for closeness / similarity metrics in an efficient and optimized search.

The invention discloses a building physical fuzzy query method based on a large language model and a medium. The method comprises the following steps: acquiring a current to-be-queried user problem in real time in a building physical fuzzy query system, and analyzing the current to-be-queried user problem to obtain an analysis result of the current to-be-queried user problem; through a multi-type hierarchical fuzzy matching method, comparing the problem analysis result of the current to-be-queried user to generate current matching query key information; inputting the current matching query key information into a pre-constructed building data dictionary knowledge graph and a building physical vector database, and generating a current information fusion candidate answer; and verifying the current information fusion candidate answer through a large language model answer matching degree verification method, and if the verification is passed, obtaining a target fuzzy query answer. The problems of low query efficiency and incompleteness caused by lack of a fuzzy query mechanism for building physical system characteristics are solved, and the building physical fuzzy query efficiency and accuracy are improved.

The present application belongs to the technical field of data retrieval, and particularly relates to a method for improving data retrieval speed, comprising the following steps: S1, generating a multi-table fuzzy query request; S2, building a business physical view service; S3, dimension reduction of the multi-table fuzzy query request, and single-table query of the business physical view; wherein the business physical view is classified into a single physical view for the same type of business function. The present application establishes a business physical view service, creates and updates the business physical view to be queried in real time, and performs full-text indexing on the business physical view, so as to provide single-physical-view fast fuzzy query for business query, make the fuzzy query not limited by left and right fuzzy queries, and greatly improve the query efficiency from tens of seconds or even hundreds of seconds to tens of milliseconds, thereby greatly improving the user experience, especially for multi-table joint query and multi-field simultaneous query, which more embodies the query advantage.

The invention discloses an engineering material price intelligent retrieval method based on fine-grained field perception fusion. The method comprises the steps that an original query set is received, and a material library is provided; performing semantic alignment and structure perception enhancement processing on the original queries in the original query set to generate standardized enhanced queries; calculating the global similarity of normalized enhanced query and records in a material library through a dense semantic channel and a sparse lexical item channel; performing field-level analysis on each record in the standardized enhanced query and material library, and calculating field-level fine-grained similarity of the dense semantic channel and the sparse lexical item channel on a predefined key field set; carrying out normalization processing on the global similarity and the field-level fine-grained similarity, and carrying out weighted fusion to generate a final fusion similarity; and sorting the records in the material library according to the final fusion similarity, and outputting the record with the highest rank as target material price information. The method effectively relieves the influence of noise and fuzzy query on the matching precision.

The invention provides a vector database construction method and system, and relates to the technical field of databases. The method comprises the steps of performing normalization processing on an original vector so as to limit a sudden change value and zoom the sudden change value to a unified range; recursive dichotomy is carried out on the value range of each dimension, binary codes are allocated to each interval, and mapping of the intervals and the codes is dynamically recorded through a skip list; alternately combining codes of all dimensions of the normalized vector to generate a global binary string; and finally, taking the binary string and the original vector as a joint primary key, and constructing a B + tree index. According to the method, a high-dimensional proximity problem is converted into a prefix matching problem of a one-dimensional space, and excellent characteristics of a B + tree and a skip list are utilized, so that efficient insertion, deletion, modification, accurate query and fuzzy query of high-dimensional vector data are realized; the problems of complicated calculation, high storage overhead, difficulty in dynamic updating and low query speed of the traditional method on high-dimensional data are effectively solved.

The invention relates to the technical field of rail transit inspection, and discloses a rail transit inspection platform based on man-machine interaction, which comprises a data acquisition module, a fuzzy label module, a label association module, an intelligent calling module, a user feedback module and a fuzzy query module. Polling information is labeled through a fuzzy label system, a correlation model is constructed to realize deep reasoning between labels, and a data processing flow is optimized in combination with intelligent calling and user feedback. According to the invention, the organization efficiency and the intelligent level of the inspection data can be improved, the manual operation complexity is reduced, and the defect identification accuracy and the inspection work reliability are improved.

The invention discloses a large model dialogue information collection method and system fused with business query. The method comprises the steps that initial input of a user is received, a session variable is initialized, a predefined cue word template is loaded and combined with a session state filling placeholder to generate a calling cue word, and a large model is called to extract a candidate structured variable; when it is judged that service query needs to be executed, multi-level fuzzy query is executed according to accurate, fuzzy and auxiliary matching priorities, a dialogue strategy is adjusted according to a result, variables are confirmed, and a clear list or a question is generated; and when the session variables are updated, converting the user anaphora and calculating the confidence of the active list to determine a target item, and generating response iteration until all target variables are extracted. Based on the construction of the system, the problems of insufficient fusion of information extraction and business verification, fuzzy reference and multi-list confusion in the prior art are solved, the variable extraction accuracy and the dialogue efficiency are improved, and the system is adaptive to complex multi-round business dialogue scenes.

The invention discloses a vector node optimization method and system for enhancing RAG retrieval accuracy, and particularly relates to the technical field of information retrieval and natural language process.The method comprises the steps that scenarized understanding and reconstruction are conducted on an original query of a user, and a target query is generated; based on the target query, calculating scene adaptive dynamic weights fusing semantic relevancy, timeliness and authority for the knowledge base document nodes; multi-channel retrieval is triggered in a self-adaptive mode according to the scene and complexity, and preliminary sorting is optimized through dynamic weight; performing multi-dimensional verification and reordering on the primary result, and outputting a high-order result; and continuously optimizing the system through feedback learning. According to the method, the problems of fuzzy query, static weight and single strategy in traditional retrieval are solved, the spanning from semantic matching to scenarized accurate matching is realized, and the accuracy, adaptability and result reliability of retrieval are remarkably improved.

The application relates to an ID fuzzy query method and device, electronic equipment and a storage medium. The method comprises the following steps: establishing an ID query tree; when a target message to be queried is received, determining the first target byte length corresponding to the ID of indefinite length in the target message; determining the first byte length group to which the first target byte length belongs from a plurality of preset byte length groups and determining the first preset byte length corresponding to the first byte length group; extracting the first N bytes from the target message, wherein N is the first preset byte length; determining the hash value of the first N bytes to obtain the target hash value corresponding to the ID of indefinite length in the target message; and according to the target hash value, querying the ID query tree and the prefix subset ID corresponding to each preset hash value to obtain the fuzzy query result corresponding to the ID of indefinite length in the target message. The application improves the efficiency of the query of the ID of indefinite length.

The invention is suitable for the technical field of information retrieval, and provides a knowledge base-based interactive retrieval method, system, device and program product, the method comprises the following steps: receiving an original query input by a user, and carrying out intention confidence detection on the original query to generate a confidence score representing the intention definition degree of the original query; comparing the confidence coefficient score with a preset confidence coefficient threshold value, if the confidence coefficient score is greater than the confidence coefficient threshold value, judging the original query as a clear query, and executing retrieval based on the clear query to obtain a first retrieval result; if the confidence score is smaller than the confidence threshold, judging the original query as a fuzzy query, and executing an interactive clarification operation; wherein the interactive clarification operation comprises the steps of determining at least one clarification dimension for clarifying the fuzzy query, generating a plurality of clarification options based on the clarification dimension, and presenting the clarification options on an interactive interface to obtain feedback input of a user; and in response to the feedback input, forming a new query in combination with the fuzzy query and the feedback input, and performing retrieval based on the new query to obtain a second retrieval result. Therefore, the accuracy and correlation of search results are improved.

The present disclosure provides a fuzzy query method, which can be applied to the field of artificial intelligence technology. The method comprises: acquiring a query statement input by a user and multi-source heterogeneous text data; interacting and matching the query statement with an external knowledge base to generate a knowledge-enhanced semantic representation vector; performing hierarchical feature fusion on the semantic representation vector and the multi-source heterogeneous text data to obtain text representation with fused knowledge features, wherein the hierarchical feature fusion comprises layer-by-layer aggregation of word-level, sentence-level and document-level features through a Transformer encoder; inputting the text representation with fused knowledge features into a multi-task learning framework for intent recognition and slot extraction to obtain a structured query understanding result; generating a time-space retrieval condition according to time entities and geographic location entities in the structured query understanding result, and outputting a corresponding retrieval analysis result. The present disclosure also provides an event cognitive reasoning device, equipment, storage medium and program product.

The invention discloses a medical knowledge query language design and semantic mapping method, and belongs to the technical field of information. Aiming at the problems that a general knowledge graph query language is poor in adaptability and high in use threshold in a medical knowledge query scene, the invention designs a domain-specific query language ClinicalDSL, a keyword system of the domain-specific query language clinicalDSL fits clinical terms, a grammar rule adopts an EBNF-like specification, and the use threshold is high. Five high-frequency clinical query types including condition query, statistical query, time sequence query, fuzzy query and set query are supported; a mapping algorithm from ClinicalDSL to a semantic model is put forward, clauses such as query conditions, targets, sorting and grouping are converted into the semantic model formed by element operations such as node selection, node positioning, time sequence operation, set operation, statistical operation and output control by recursively traversing a grammar analysis tree from top to bottom, and decoupling with a bottom layer graph database is achieved. According to the method, the diagnosis and treatment intention can be accurately expressed without mastering a complex query language, and the availability and efficiency of the electronic medical record knowledge graph in query are improved.

The invention provides a fuzzy query method which can be applied to the technical field of artificial intelligence. The method comprises the steps of obtaining a query statement and multi-source heterogeneous text data input by a user; performing interactive matching on the query statement and an external knowledge base to generate a knowledge-enhanced semantic representation vector; hierarchical feature fusion is carried out on the semantic representation vector and the multi-source heterogeneous text data to obtain text representation fused with knowledge features, and the hierarchical feature fusion comprises the step of aggregating word-level, sentence-level and document-level features layer by layer through a Transform encoder; the text representation fused with the knowledge features is input into a multi-task learning framework for intention recognition and slot extraction, and a structured query understanding result is obtained; and according to the time entity and the geographic position entity in the structured query understanding result, generating a space-time retrieval condition, and outputting a corresponding retrieval analysis result. The invention further provides an event cognitive reasoning device and equipment, a storage medium and a program product.

The application relates to the computer technical field, in particular to a data fuzzy matching query method and device, electronic equipment and storage medium, wherein the method comprises the following steps: receiving a to-be-queried parameter, and performing data type conversion on the to-be-queried parameter to obtain a to-be-queried array; judging whether the to-be-queried array exists in a preset data structure table; if the to-be-queried array exists in the preset data structure table, generating a query result based on the content corresponding to the to-be-queried array, and returning the query result to a preset user end. Therefore, the problems that some fuzzy query methods in the related art are slow in index speed and low in accuracy, some methods are high in hardware resource requirement and high in cost are solved, the speed of fuzzy matching query is improved, the accuracy of matching is effectively improved, and the method has the characteristics of being simple, easy to use and low in cost.

The invention relates to an information search method and device, electronic equipment and a storage medium, and the method comprises the steps: recognizing a search intention of search information under the condition that the search information is received; wherein the search intention is used for representing fuzzy query in search information; according to the search intention, determining an associated knowledge graph associated with the search intention from a pre-constructed object knowledge graph; processing the search information according to the associated knowledge graph to obtain target search information; and searching according to the target search information to obtain a search result. Therefore, the accuracy of information search can be improved.

The embodiment of the invention discloses a fuzzy query method, system and device, a medium and a product, and the method comprises the steps: carrying out the encryption mapping of characters of plaintext data based on a dynamic disturbance factor, and generating a coding sequence according to the encryption mapping result; constructing a reverse index based on the coding sequence; performing encryption mapping on characters of a query character string based on the dynamic perturbation factor to generate query segmented words; searching the query segmented word according to the inverted index to obtain a candidate result set containing position information corresponding to the query segmented word; and performing sequential semantic verification on the candidate result set according to the position information to obtain a fuzzy query result. According to the method and the device, high-efficiency encrypted field fuzzy retrieval is realized while the security of the stored sensitive field data is ensured.

The application discloses a kind of through signal intelligent matching spacecraft electrical connection relationship design method, which comprises three steps of automatic matching connector, manual matching, fuzzy query connector: automatic matching: the definition information of equipment and connector input by interface data, automatically find signal matching connector to connect, and establish connection relationship if connection is successful;Manual matching: using interface input and the link relationship information established by automatic matching, display all devices and connectors in the system in tree menu, as well as the state of connection relationship;Intelligent query: in the process of manually matching connector, according to the node information of connector, provide a visual graphical interface, and list all connectors that meet the matching conditions and the pins they belong to.The application uses information technology to realize electrical connector data, with automatic matching as the main method and manual operation as the auxiliary method, to automatically establish link relationship and achieve the purpose of automatically generating cable data.

The invention discloses a sensitive data ciphertext fuzzy query method and device, equipment, a medium and a product. According to a query target, a primary key ID of a Bloom filter which is encrypted in advance and stores different characters in sensitive data is obtained; splitting the query target into a plurality of character strings spliced with the primary key ID; and comparing the plurality of character strings with data stored in different Bloom filters one by one, and determining a query result. According to the scheme, fuzzy query of any mobile phone number group can be realized on the premise of additionally increasing a small storage space.