Patents
Literature
8results about How to "Reduce search" 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
A stock ownership penetration method based on bidirectional search, a terminal and a storage medium
ActiveCN121326984BImprove discovery efficiencyNarrow down the search space
The application relates to the technical field of data storage and processing, and discloses a stock right penetration method based on bidirectional search, a terminal and a storage medium. The method constructs a stock right graph structure: based on original stock holding data, a directed graph is constructed; a vertex set represents an institution entity, and an edge set represents a stock holding relationship between institutions; the weight of an edge in the stock right graph represents a stock holding proportion between institutions, and a time stamp information is additionally attached to the edge to record the timeliness of the stock holding relationship; graph structure cleaning: deleting an unweighted edge in the stock right graph, detecting and eliminating a circular stock holding; executing bidirectional penetration search: for any given starting point and ending point in the stock right graph, initiating a forward search from the starting point to obtain a forward path, initiating a reverse search from the ending point to obtain a reverse path, and when the forward search and the reverse search meet at a vertex, the reverse path of the forward path and the reverse path are combined, so that a complete stock holding link is obtained. The application can effectively realize stock right penetration between institutions.
Owner:HEFEI DAZHIHUI CAIHUI DATA TECH CO LTD
Intelligent analysis and recommendation method for bidding documents, computer device and readable storage medium
PendingCN122173492Areduce searchimprove accuracyDatabase queryingDatabase modelsEngineeringKnowledge graph
This application discloses a method, computer device, and readable storage medium for intelligent analysis and recommendation of tender documents, belonging to the field of artificial intelligence and natural language processing technology. The method includes the following steps: constructing a knowledge graph based on acquired target content related to tendering and bidding, and calling an embedding model to generate text vectors; storing the knowledge graph in a graph database and the text vectors in a vector database; responding to a user's query request, extracting low-level keywords and high-level keywords based on the query request; wherein the query request carries at least the tender information of the tender project; performing graph vector collaborative retrieval in the graph database and the vector database based on the low-level keywords and the high-level keywords to obtain a multi-source fusion context; and generating a tender recommendation scheme based on the multi-source fusion context. This application can improve the accuracy of tendering and bidding analysis.
Owner:NORTH CHINA UNIV OF WATER RESOURCES & ELECTRIC POWER
Medical data retrieval method and system based on event-driven execution chain, medium, program product and terminal
PendingCN121833806AImprove adaptabilityreduce searchMedical data miningDigital data information retrievalData setComputational logic
The invention provides a medical data retrieval method and system based on an event-driven execution chain, a medium, a program product and a terminal. According to the method, the retrieval method with the patient-level medical event time sequence as the execution object and the execution sequence of the event-driven execution chain as the core is constructed, the retrieval method is different from a retrieval mode based on rule configuration or event display in the prior art, innovation is achieved from the execution mechanism level, and the adaptability of a retrieval scheme to medical event time sequence characteristics is effectively improved. According to the method, the computing node for cross-event index computing is used as the propelling condition of the event-driven execution chain, so that the computing logic is fused into the core link of the retrieval process instead of being used as an independent post-event analysis step, the unnecessary retrieval process is reduced, and the retrieval period is shortened. Through the decoupling design of the event-driven execution chain and the patient data, the same event-driven execution chain can be reused across different data sets and different research items, repeated development work is reduced, and the technical landing cost of clinical research is reduced.
Owner:SHANGHAI NAT GRP HEALTH TECH CO LTD
Course scheduling optimization method and apparatus executed by electronic device
PendingCN121961795Aincrease probabilityReduce the difficulty of solvingForecastingComplex mathematical operationsLocal optimumAlgorithm
The invention provides a course scheduling optimization method and device executed by electronic equipment, and can be applied to the technical field of computer technologies. The method comprises the following steps: constructing a cumulative retention function of a user through an arithmetic unit based on scheduling data of courses and preset user experience data; decomposing the cumulative retention function to obtain a plurality of sub cumulative retention functions; adding the initial solution space into a to-be-understood space list, and storing the to-be-understood space list into a register; n times of solving are carried out based on the initial solution space, and the nth time of solving comprises the steps that the current solution space is solved based on a plurality of sub cumulative retention functions through a solver, and a local optimal solution and a sub cumulative retention value of the current solution space are obtained; under the condition that the local optimal solution is located in the current solution space, dividing the current solution space based on the local optimal solution to obtain a plurality of sub-solution spaces; and determining an optimal curriculum arrangement strategy based on the local optimal solution, and scheduling the curriculums based on the optimal curriculum arrangement strategy.
Owner:UNIV OF SCI & TECH OF CHINA
Adaptive dynamic optimization method and system for retrieval threshold of RAG knowledge base
PendingCN122262275AImprove recallnoise controlDigital data information retrievalBiological modelsShardStatistical analysis
The application provides a kind of RAG knowledge base retrieval threshold self-adapting dynamic optimization method and system thereof, comprising reading system basic configuration parameter, and based on parameter dynamic calculation maximum shard containing number;Load test question set, execute threshold step scanning to generate candidate threshold;For each candidate threshold, retrieve test question, record the original data of retrieval result under each candidate threshold;Statistical analysis is carried out on the retrieval result, and effective coverage rate and average result number are calculated;Apply constraint-based nonlinear scoring algorithm, combine effective coverage rate and average result number, and introduce load adaptability as penalty boundary, calculate the comprehensive score of each candidate threshold;According to the comprehensive score result, output the threshold with the highest score as the recommended parameter, and generate an analysis report containing the recommended reason.The application effectively controls the context noise and Token cost input to the large model while ensuring the retrieval recall rate, and realizes the automatic performance optimization of RAG system.
Owner:ZHUHAI GOTECH INTELLIGENT TECH CO LTD
Large language model reasoning methods, systems, and storage media based on relational constraints and path guidance
PendingCN122088689Areduce distractionsReduce hallucinationsKnowledge representationInference methodsLinguistic modelAlgorithm
This invention relates to a large language model reasoning method based on relational constraints and path guidance. The method receives an input question through a large language model and generates a set of relational constraints semantically related to the input question by combining preset prompts. Using a knowledge graph as the retrieval object, multi-hop relation retrieval is performed iteratively based on the generated set of relational constraints. After each hop, the semantic similarity between candidate relations and the input question and relational constraints is calculated, constructing a constrained multi-hop relation space. The multi-hop relation space is combined with the input question, and an interpretable relational path is generated through the large language model. Based on the generated relational path, corresponding candidate subgraphs are retrieved from the knowledge graph. Through hop-by-hop entity filtering and pruning operations, the relational path is instantiated into an entity-level reasoning path. The obtained reasoning path is combined with the input question, and an accurate and interpretable answer is generated through the large language model.
Owner:TIANJIN NORMAL UNIVERSITY
A stereoscopic storage mechanism
Owner:JIAYUGUAN SHUNHENG COLD CHAIN DEV CO LTD
A method for optimizing parameters of a PID controller of a water turbine speed regulation system
PendingCN122359214AImprove adaptabilityImprove anti-interference abilityWater turbineMathematical model
The application relates to a PID controller parameter optimization method of a water turbine speed regulation system, which comprises the following steps: according to a parallel PID controller structure, a three-order nonlinear object is adopted to establish a water turbine speed regulation system mathematical model; according to the water turbine speed regulation system mathematical model, an optimization objective function of the water turbine speed regulation system under no-load working conditions or load working conditions is established; according to the optimization objective function and the constraint range of the PID controller parameters, an improved sparrow search algorithm is adopted to optimize the PID controller parameters, and optimal PID controller parameters are output, wherein the improved sparrow search algorithm introduces a water hammer coefficient in the producer search step length of a basic sparrow search algorithm and adopts boundary adsorption combined with gradient correction instead of random disturbance. The application realizes global optimization and engineering adaptation of PID parameters, and effectively improves the control performance and operation stability of the speed regulation system under all working conditions.
Owner:HARBIN ENG UNIV