Patents
Literature
24results about How to "Reduce retrieval time" patented technology
Efficacy Topic
Property
Owner
Technical Advancement
Application Domain
Technology Topic
Technology Field Word
Patent Country/Region
Patent Type
Patent Status
Application Year
Inventor
Document question and answer method and system based on LLM, RAG and knowledge graph
InactiveCN119474307AImprove efficiencyreduce retrieval timeSemantic analysisKnowledge representationQuestions and answersKnowledge graph
The invention discloses a document question and answer method and system based on LLM, RAG and a knowledge graph. The method comprises the steps that an obtained historical question set and document paragraphs related to the historical question set are classified; extracting a plurality of first keywords in each type of historical problem document set; utilizing a knowledge graph to construct a relationship between the plurality of first keywords and the document paragraphs related to the plurality of first keywords to obtain a relationship knowledge graph; constructing a keyword library containing a plurality of first keywords and words related to the first keywords; extracting a plurality of second keywords in the to-be-answered question, and searching a plurality of third keywords similar to the plurality of second keywords from a keyword library; retrieving a plurality of document paragraphs corresponding to the plurality of second keywords and the plurality of third keywords from a relational knowledge graph; and inputting the to-be-answered question and the plurality of document paragraphs into the trained prediction large model to obtain a target answer. The accuracy and efficiency of document question answering can be improved.
Owner:GUANGXI COMM GUIHUA DESIGN CONSULTATION CO LTD
Dynamic module operation management system based on programming language
PendingCN119248255Areduce retrieval timereduce wasteProgram code adaptionProgram/content distribution protectionSoftware engineeringData access
The invention discloses a dynamic module operation management system based on a programming language, and the system comprises a module analyzer which manages the dependency relationship between modules in the system, applies a compiling optimization technology before the modules are loaded, and optimizes the code execution through code expansion and constant folding; operation environment establishment: optimizing execution speed and resource scheduling by integrating a just-in-time compiling JIT technology and an optimization measure during operation; the module warehouse is used for controlling the retrieval and loading process of the module by adopting an optimized organization and index strategy; data structure optimization: using a skip list to adjust data structure design, and controlling data processing efficiency; dynamic optimization: the system dynamically adjusts an optimization strategy according to performance feedback and an operation environment, and controls improvement of self-adaptive performance; garbage collection: dynamically adjusting a garbage collection strategy by adopting a self-adaptive garbage collection algorithm, and reducing the module retrieval time, accelerating the module loading process and improving the data access efficiency.
Owner:SITEWEI (SUZHOU) TECH CO LTD
Data security document generation method and system based on RAG technology
InactiveCN119475436AImprove knowledge reserveclosely guardedDigital data information retrievalDigital data protectionData OriginData source
The invention discloses a data security document generation method and system based on an RAG technology, and relates to the technical field of artificial intelligence, and the method comprises the steps: carrying out isolation processing on data security related data based on data source information; storing the encrypted data and the unencrypted private data into a private data security database, and updating the mass data security vertical industry data into the private data security database in real time; in response to a generation request of the data security document, extracting a keyword corresponding to a user demand, and performing similarity retrieval in the private data security database based on the keyword to obtain corpus data having the highest similarity with the keyword; and the corpus data, the user demand and the final document template are jointly input into the large language model to generate the data security document, so that the efficiency and accuracy of generating the data security document are improved.
Owner:BEIJING LIANSHI NETWORKS TECH CO LTD
Construction method and system of folk song inheritance atlas
InactiveCN119539062Aefficient updateimprove protectionKnowledge representationSpecial data processing applicationsAlgorithmTheoretical computer science
The invention discloses a method and a system for constructing a folk song inheritance atlas. The method comprises the following steps: acquiring lyric song text data of a folk song sample; dividing the subordinate relationship hierarchical structure of the folk song according to the text description; on the basis of the divided subordinate relationship hierarchical structure of the folk songs, representing the folk song samples as triads, and storing the triads into a database; generating a civil song sample tree based on the obtained civil song sample triple data; based on nodes with non-final knowledge in the civil song sample tree, performing reforming construction on the civil song sample tree to obtain a complete civil song sample tree; the method for constructing the civil song inheritance atlas based on the triple has the advantages that the civil song sample tree is constructed and the query efficiency is optimized through the digitalized and structured data model, and the systematic storage and inheritance efficiency of civil songs is effectively improved.
Owner:XICHANG COLLEGE
Logo retrieval method and device, computer equipment and storage medium
PendingCN119474425Areduce retrieval timeimprove accuracyDigital data information retrievalMachine learningCosine similarityFeature vector
The invention belongs to the field of artificial intelligence, and relates to a l ogo retrieval method, which comprises the following steps of: pre-processing all l ogo images in a l ogo material library in advance to obtain a material image; rotating the material image at different angles, and performing feature extraction on the material image at different rotation angles through a preset multi-modal model to obtain feature vectors of the material image at different rotation angles; combining all the feature vectors as index vectors of the corresponding material images, and constructing a vector database based on the index vectors; obtaining a query request, and extracting text features of the query request to obtain a query vector; and calculating the cosine similarity between the query vector and each index vector in the vector database, and determining a query result according to the cosine similarity. The invention further provides a logo retrieval device, computer equipment and a storage medium. According to the method, the index vector of the material image in the logo material library is determined in advance, and only the text vector of the query request needs to be calculated during each retrieval, so that the retrieval time is greatly shortened.
Owner:PING AN BANK CO LTD
A method for improving cable insulation performance based on production process parameter adjustment
InactiveCN118692752Bincrease risk resistanceReduce operating pressureDesign optimisation/simulationSpecial data processing applicationsProcess engineeringElectric cables
Owner:HEFEI UNIV OF TECH +1
Vision-based method and system for retrieving lost property along railway
ActiveCN119540873AEfficient use ofImprove retrieval efficiencyDigital data information retrievalBiological modelsFeature extractionVision based
The invention belongs to the field of electric data processing, and provides a vision-based method and system for retrieving lost property along a railway, which comprises the following steps: acquiring a monitoring video along the railway, framing to obtain a search library, and filtering all images and lost property images in the search library to obtain a filtered image search library and lost property filtered images; based on the filtered image search library and the lost property filtered image, performing feature extraction by using the trained double-branch convolutional neural network model to obtain a search feature library and lost property features; respectively extracting binary search library image features of the search feature library and binary lost property image features of the lost property features by calculating difference value indexes and fluctuation value indexes of all features in the search feature library and the lost property features, and calculating the matching degree of the binary search library image features and the binary lost property image features; outputting a retrieval result according to the matching degree; the defects of a vision-based method for retrieving the lost property along the railway are overcome.
Owner:LANZHOU JIAOTONG UNIV
Operation method and device of AI question-answering system, equipment and medium
PendingCN122087080Areduce retrieval timehigh speedSemantic analysisInference methodsUser inputEngineering
The invention discloses an operation method and device of an AI question-answering system, equipment and a medium, and aims to solve the problem that the accuracy of answers answered by the AI question-answering system is not high due to the fact that a traditional AI question-answering system usually adopts a mode that questions are matched with documents. Comprising the steps of converting a natural language query request input by a user into a query vector; performing vector similarity matching in a problem unit based on the query vector; and if the vector similarity matching is hit, obtaining a hit information unit associated with the question unit, and returning the hit information unit to the user as a standard answer. According to the AI question answering system, large model illusion can be reduced, answers are more accurate, retrieval time is shortened, and the speed of obtaining the answers is higher.
Owner:SUZHOU YULING ARTIFICIAL INTELLIGENCE TECHNOLOGY CO LTD
WBS list intelligent pushing and performance tracking method based on experience case correlation
The application provides a WBS list intelligent pushing and performance tracking method based on experience system correlation, relates to the technical field of project management and intelligent algorithm, and comprises the following steps: constructing a multilevel three-dimensional WBS correlation library, building a correlation system of business blocks and WBS nodes based on unified coding rules, establishing a three-dimensional correlation relationship among resources, WBS and project scenarios, inputting multiple types of resources and binding WBS node information and correlation attributes, and realizing the configurability of correlation relationship and resource priority and the dynamic updating of the correlation library; the application solves the problems of WBS correlation resource fragmentation, low correlation degree and update lag in the prior art, realizes the deep binding of resources, WBS and project scenarios through a multilevel three-dimensional WBS correlation library construction mechanism, combines correlation strength quantitative calculation and dynamic updating capability, makes resource reuse more accurate and data support more solid, improves resource utilization efficiency, shortens resource retrieval time, and improves resource matching accuracy.
Owner:SHANDONG ELECTRIC POWER CONSTR NO 2
A hospital internal audit data processing system and method
ActiveCN118412106Bease of workincrease flexibilityFinanceDigital data protectionData processing systemData acquisition
The present invention discloses a hospital internal audit data processing system and method. According to the technical solution provided by the present invention, the system includes: an audit plan module, an audit preparation module, an audit implementation module, an audit report module and an audit rectification module; the audit plan module is used to prepare various types of audit plans, establish and manage audit projects; the audit preparation module is used to manage pre-audit matters of audit projects; the audit implementation module is used to perform data analysis and manuscript management during the audit process, and obtain audit results; the audit report module is used to feedback audit results and generate audit reports; the audit rectification module is used to send audit rectification notices, obtain feedback on rectification, and identify rectification. Through the present invention, different audit plans and audit projects are created according to the demand type, and data is sorted or encrypted, which enhances data standardization and transmission security, completes the integration from data collection to report sending and automatic archiving, and improves efficiency.
Owner:SHENZHEN TRADITIONAL CHINESE MEDICINE HOSPITAL
Medical retrieval system based on semantic logic
PendingCN119474412AHigh precisionreduce retrieval timeSemantic analysisMedical referencesSearch wordsMedical statistics
The invention is applicable to the technical field of intelligent medical treatment, and provides a semantic logic-based medical retrieval system, which comprises an acquisition module, a construction module, a semantic analysis module, a processing module, a storage module and a retrieval module, the construction module is used for constructing a semantic word bank by the words acquired from the target data source by the acquisition module; the semantic analysis module is used for classifying words in the constructed semantic word bank through semantic analysis; the storage module is used for storing the words analyzed and classified by the semantic analysis module into a storage unit; the processing module is used for determining a keyword in a to-be-processed statement and determining a data source within a certain range; when the system is used for searching the search word, the search is only carried out in the literature containing the medical statistics method, so that the accuracy of the search result is improved, and the search word and the word with the same semantics or association relationship with the search word are used for searching, namely, the search range is expanded, and the recall ratio is improved.
Owner:CHENGDU KNOWLEDGE VISION SCI & TECH CO LTD
Clothing rack for artistic design
ActiveCN224474246UBreak through the limitations of a single dimensionreduce retrieval timeControl engineeringDust control
The utility model relates to clothing rack technical field, and disclose a clothing rack for art design, including base and top frame, and the base is installed with a electric telescopic column between top frame, and the base is rotatably installed with the drive of the main drive swing frame through the transfer position motor, and the main drive swing frame is rotatably installed with the drive of the transmission bevel gear through servo motor, and the top frame is rotatably installed with driven swing frame, and a group of hanger module is installed between the main drive swing frame and driven swing frame, and the hanger module includes elastic ring carrier tape. The utility model passes through the multidimensional cooperation of mechanical drive, intelligent control and material innovation, and the core problem such as traditional clothing rack display angle limited, form rigid, dust difficult to clean and operation inefficiency is systematically solved. Its technical scheme has remarkable creativity in three-dimensional display, dynamic form adjustment, automatic dust removal and remote control etc.
Owner:ANHUI YISHANG CLOTHING CO LTD
Vehicle information retrieval method, device, electronic device and storage medium
ActiveCN113961671Bquick searchreduce retrieval timeDigital data information retrievalNatural language data processing
The present application relates to the field of vehicle information technology, and in particular to a vehicle information retrieval method, device, electronic device and storage medium. The retrieval method comprises: obtaining a retrieval request of a target vehicle, wherein the retrieval request includes a fuzzy license plate number of the target vehicle; determining a target fuzzy type of the fuzzy license plate number according to the number of digits and positions of fuzzy characters in the fuzzy license plate number; after determining that the license plate number segmentation type in the license plate number retrieval library includes the target fuzzy type, the precise license plate number corresponding to the target vehicle is retrieved from the license plate retrieval library according to the fuzzy license plate number and the target fuzzy type. The present application determines the license plate number segmentation type in the license plate number retrieval library in advance, so that after obtaining the retrieval request of the target vehicle, the precise license plate number corresponding to the target vehicle can be quickly retrieved in the license plate number retrieval library, thereby shortening the retrieval time and reducing the retrieval scope, thereby improving the retrieval efficiency.
Owner:ZHEJIANG LIYUAN TECH CO LTD
A context-aware dialogue recommendation method, system, device and storage medium
PendingCN122734019AAvoid false triggering of functionsHigh trigger reliability
The application provides a context-aware dialogue recommendation method, system, device and storage medium, and belongs to the technical field of intelligent customer service, and specifically comprises receiving a trigger request, acquiring a current desktop image, identifying a new message, judging an overlap degree comparison mode of the desktop image and the bottom message, when it is determined that it is a real-time dialogue mode, identifying a question, judging whether the question exists in a historical snapshot cache, when it is determined that the question exists, acquiring a historical snapshot corresponding to the question, when it is determined that the question does not exist, triggering a double-channel asynchronous retrieval, realizing double-channel asynchronous retrieval through vector retrieval and large language model retrieval in a dialogue knowledge base, acquiring a retrieval result corresponding to the question, storing the retrieval result and the question as a new historical snapshot cache to provide an exhibition interface of a customer service terminal. Common questions are reused by snapshot cache, data calculation and retrieval time consumption are reduced, a large language model retrieval result is used to cover a vector retrieval result, and it is ensured that the retrieval result is more matched.
Owner:GUANGDIAN EMPOWERMENT TECHNOLOGY (SHANGHAI) CO LTD
Vector library pre-merging method for improving vector retrieval speed
PendingCN119669272AImprove retrieval speedreduce dispersionEnergy efficient computingSpecial data processing applicationsData acquisitionEngineering
The invention discloses a method for improving the vector retrieval speed through vector library pre-merging. The method comprises the following steps of 1, data collection and preprocessing; 2, real-time vector library division: dividing the preprocessed vector data into a plurality of sub-vector libraries according to a certain rule; step 3, sub-vector library index construction: constructing an inverted index by taking the three-level sub-vector library as granularity; step 4, performing vector library pre-merging according to the performing retrieval through an inverted index according to a business demand expression provided by an upper-layer actual business application; 5, monitoring and updating pre-merging trigger condition judgment; and step 6, performing optimization processing after pre-merging. According to the method, by pre-merging and optimizing the vector library, the dispersion and redundancy of the data are reduced, the indexing efficiency is improved, and the retrieval time is greatly shortened.
Owner:UESTC COMSYS INFORMATION
An accounting voucher sorting device that facilitates data classification
ActiveCN224617254UReduce alignment and finishing burdenReduce repetitive labor burden
This utility model provides an accounting voucher sorting device that facilitates data classification, relating to the field of accounting equipment technology. It includes a voucher storage box, a primary connecting bracket, and a secondary connecting bracket. The primary connecting bracket is rotatably connected to the primary connecting bracket at its primary end. An upper threaded connecting frame is threaded to the upper part of the secondary connecting bracket, and a top sliding protrusion is fixedly connected to the top of the upper threaded connecting frame. A bottom sliding groove is formed at the bottom of the secondary connecting bracket, and the bottom sliding groove of the secondary connecting bracket slides into and connects to the top sliding protrusion of the upper threaded connecting frame of another secondary connecting bracket. The secondary connecting brackets are vertically stacked through the sliding of the bottom sliding groove and the top sliding protrusion. The stacking of the voucher storage boxes improves space utilization, reduces space occupation, and shortens the classification and sorting time, solving the problem of insufficient space due to limited desktop space in planar classification methods.
Owner:INNER MONGOLIA FINANCE AND ECONOMICS UNIVERSITY
A fast strategy retrieval method based on attribute bloom filter
ActiveCN114268482Bachieve retrievalAddress privacy breachesKey distribution for secure communicationUser identity/authority verificationHash functionCiphertext
The present invention discloses a fast policy retrieval method based on attribute Bloom filter. The data owner generates the parameters required for platform operation and generates the available private key of the data user by using the subject attributes. The data owner formulates the access control policy and converts the policy into a linear secret sharing matrix to realize the construction of the policy and publish it to the blockchain. Before knowing the message to be encrypted, the data owner performs a large number of pre-calculations based on the public parameters. When the message to be encrypted is known, the ciphertext is quickly generated. During policy retrieval, the attribute matching function is reconstructed by using the attribute Bloom filter, and the specific position of the subject attribute is quickly located by using the hash function. If the attribute mapping function can be successfully constructed, it proves that the attribute of the data user meets the access policy formulated by the data owner, otherwise it indicates that there is no permission to access. Finally, the address of the object resource is decrypted, and the pressure of the chain retrieval is dispersed through multiple clients, the system retrieval time is improved, and data sharing is completed.
Owner:BEIJING UNIV OF TECH
Large model-based structured xml standard document intelligent query assistance system
ActiveCN121434158BImprove query efficiencyavoid missingFile metadata searchingFile/folder operationsEngineeringData profiling
The application relates to the technical field of document query assistance, in particular to a structured XML standard document intelligent query assistance system based on a large model, which comprises a data receiving module, a document comparison module, a data division module, a data analysis module, a text generation module, a data statistics module and a data processing module, whether the processing of each document is qualified is determined based on a click probability, when it is determined that the processing of each document is abnormal, a preset reference quantity is adjusted to a corresponding value, whether the operation of the document generation module is qualified is determined based on the average push quantity of the average single output push document of the text generation module, when it is determined that the operation of the document generation module is abnormal, a preset category quantity used for dividing the relevance of single batch documents is adjusted to a corresponding value. The feedback of the user is analyzed, the system is optimized according to the use behavior of the user, and the query efficiency of the document is improved.
Owner:CHINA NAT INST OF STANDARDIZATION
A computer data integration management system and method based on the Internet
ActiveCN116340302Breduce retrieval timereduce forgettingDigital data information retrievalHardware monitoringData informationIdentity recognition
The present invention relates to the technical field of computer data processing. Specifically, it is a computer data integration management system and method based on the Internet, the computer data integration management system includes a data acquisition module, a data transmission module, a data analysis module and a data integration module; the data acquisition module collects data information on the number of transmission links in the data source front end, the data source log and the data information generated by different identity identification numbers of users; the data transmission module transmits the data information corresponding to the identity identification numbers that need to be integrated; the data analysis module analyzes the identity identification numbers of the same household in the data source and analyzes the importance of the identity identification numbers; the data integration module screens the identity identification numbers that need data integration. The present invention reduces the loss or forgetting of data information by analyzing and integrating the data of different identity identification numbers of users.
Owner:YEEPAY CO LTD
On-chain approximate nearest neighbor data privacy retrieval method and system based on locality sensitive hashing
PendingCN119272326Areduce retrieval timeImprove recallDatabase distribution/replicationDigital data protectionCiphertextData retrieval
The invention relates to the technical field of block chain data management, in particular to an on-chain approximate nearest neighbor data privacy retrieval method and system based on locality sensitive hashing. A data management node obtains data from a chain and constructs a hash table for the on-chain data by using locality sensitive hashing; encrypting the hash table into a ciphertext by using attribute-based encryption according to an access control strategy, and storing and chaining the ciphertext; and acquiring the ciphertext under the condition that the access control strategy is met, and performing on-chain data retrieval on the retrieval vector by using the smart contract to acquire on-chain data most similar to the retrieval vector. According to the method, the local sensitive hashing technology and the distributed point function are used, the private information retrieval thought and the local sensitive hashing are combined to achieve approximate nearest neighbor retrieval of the on-chain data, feasibility is provided for diversity of on-chain data management, retrieval efficiency and privacy protection can be taken into consideration, and the method is suitable for large-scale popularization and application. Efficient, accurate and safe on-chain data approximate nearest neighbor privacy retrieval can be realized.
Owner:Chinese People's Liberation Army Cyberspace Force Information Engineering University
Silent user data access method and system using ice pool
PendingCN121880271Aaccurate identificationavoid wastingFile system administrationEnergy efficient computingAccess methodIce storage
The invention relates to the field of data storage, and provides a silent user data access method and system using an ice pool. The method comprises the steps of performing activeness analysis on user operation records according to user behavior logs, performing screening to obtain a silent user set, and generating a frozen user library; performing hierarchical storage on the user accounts in the frozen user library according to freezing duration to obtain a multi-level account library; target users meeting migration conditions are extracted from the multi-level account library, file information of the multiple target users is summarized, and a user file migration list is generated; and migrating file data in the user file migration list from a hot storage pool to an ice storage pool.
Owner:E-SURFING DIGITAL LIFE TECH CO LTD
Communication fault intelligent study and judgment and artificial intelligence auxiliary emergency disposal method
PendingCN121984835Aavoid secondary failureavoid congestionDesign optimisation/simulationTransmissionFault handlingReliability engineering
The invention discloses a communication fault intelligent research and judgment and artificial intelligence auxiliary emergency disposal method, relates to the technical field of communication fault disposal, and aims to solve the technical problem that a secondary fault is easily caused due to lack of a plan verification mechanism in current traditional communication fault disposal, and the method comprises the following steps: S300, AI auxiliary decision making and plan generation; s301, combining a historical fault case library; s302, establishing a high-fidelity network digital twin body; and S303, carrying out scientific sorting by using a TOPSIS multi-attribute decision algorithm, and generating a recommended disposal plan. According to the method, the network real-time state in the current fault scene can be completely duplicated by establishing the high-fidelity network digital twinborn body, the network state change after each candidate plan is executed can be simulated in combination with the network state equation, it is ensured that the finally executed scheme has both effectiveness and safety, and the method is suitable for large-scale popularization and application. The problem that a secondary fault is easily caused due to lack of a plan verification mechanism in current traditional communication fault handling is solved.
Owner:STATE GRID HEBEI ELECTRIC POWER CO LTD
Standard layout analysis and structured storage method based on intelligent identification
PendingCN119540981AReduce storage consumptionimprove securityCharacter and pattern recognitionNeural learning methodsLayoutEngineering
The invention is suitable for the technical field of layout analysis and storage methods, and provides a standard layout analysis and structured storage method based on intelligent identification, which comprises the following operation steps: S1, data labeling: converting a standard pdf document related to the safety and reliability standard of a robot into a picture form, and carrying out binarization processing on the picture form to obtain a data label; removing the watermark of the image, and performing data annotation on the image to obtain a standard image; s2, coordinate information: carrying out layout analysis on the standard image obtained in S1 through a layout analysis network of multi-scale information fusion. According to the standard layout analysis and structured storage method based on intelligent identification, the retrieval time of the content of the robot standard document is shortened, directory structure information is effectively restored, the storage consumption of the document is reduced, the reading efficiency of the document is improved, a user can conveniently obtain the text information of the robot standard document, and the reading efficiency is improved.
Owner:CHANGZHOU INST OF INSPECTION & TESTING STANDARDS CERTIFICATION
Public health data retrieval method and system based on natural language processing
PendingCN121901242ALower the search thresholdHigh popularityDatabase management systemsSpecial data processing applicationsLinguistic modelMedicine
The invention relates to the technical field of public health information. The invention discloses a public health data retrieval method based on natural language processing. The public health data retrieval method comprises the steps that S1, a natural language retrieval instruction of a user is received; s2, analyzing an instruction through a large language model of the semantic analysis component, generating a vector value, retrieving metadata and an index component, and generating a retrieval expression; s3, checking and adjusting the expression through a semantic analysis confirmation component; s4, the data retrieval code generation component determines table attributes and association according to the expression, and SQL is generated; and S5, calling the SQL query database by the data retrieval execution component, and returning a packaging result to the user. By means of natural language processing and multi-component cooperation, the public health data retrieval threshold is lowered, and the accuracy is improved. Users can retrieve through daily languages, and non-technical users can efficiently obtain data. The system confirms a component multi-dimensional verification retrieval expression through semantic analysis, and ensures that data is matched with requirements.
Owner:ZHEJIANG CENT FOR DISEASE CONTROL & PREVENTION