Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

72results about How to "Improve logic" patented technology

Method, system and equipment for retrieval enhancement generation based on large language model and medium

The invention discloses a retrieval enhancement generation method, system and device based on a large language model and a medium. The method comprises the steps that a text file is fragmented to obtain a plurality of parent text blocks, a unique identifier is generated for each parent text block, and all the parent text blocks and the corresponding unique identifiers are stored in a relational database; fragmenting each parent text block again to obtain a plurality of child text blocks, and performing text summarization on each parent text block to obtain a plurality of parent text block summarized texts; generating one or more hypothetical questions for each parent text block; vectorizing the sub-text blocks, the summary text of the parent text block and the hypothetical question of the parent text block, and storing in a vector database; querying in the vector database and extracting the unique identifier based on a query problem text input by a user to obtain a plurality of parent text blocks; and forming prompt words by the plurality of parent text blocks and questions proposed by the user, and generating a final answer. The accuracy and recall rate of semantic retrieval are improved.
Owner:NO 30 INST OF CHINA ELECTRONIC TECH GRP CORP

Automatic prompt construction method based on man-machine conversation history and semantic retrieval

The invention relates to the technical field of artificial intelligence, and discloses a man-machine dialogue history and semantic retrieval-based prompt automatic construction method, which comprises the following steps of S1, performing text preprocessing on historical dialogue data and user input, including dynamically generating query keywords, verifying and analyzing a file uploaded by a user, and extracting core contents; s2, utilizing a pre-training language model to map the historical dialogue and the user request into semantic vectors, matching in combination with the keyword weight, and constructing a historical vector set and a user request vector; and S3, extracting a context fragment most relevant to the user request from the historical dialogue vector based on a convex optimization method. The context information highly related to the user request is accurately extracted through the sparse optimization technology, and the background information, the user request and the generation instruction are organically integrated in combination with the dynamic Prompt template design, so that the semantic correlation and context coherence of the generated reply are improved, and the interference of redundancy and irrelevant information is avoided.
Owner:BEIJING QIBU TIANXIA TECH CO LTD

Three-dimensional space positioning and path planning method of unmanned aerial vehicle pole tower

The invention relates to the technical field of space positioning and path planning, in particular to a three-dimensional space positioning and path planning method for an unmanned aerial vehicle tower. The method comprises the following steps of performing accurate longitude and latitude height data acquisition based on a laser range finder to generate accurate longitude and latitude height data; obtaining tower equipment parameters; performing inspection point three-dimensional coordinate calculation according to the tower equipment parameters and the tower three-dimensional coordinate system to generate inspection point three-dimensional coordinate data; cruise point comprehensive constraint range data are obtained according to the unmanned aerial vehicle flight association data and the polling area terrain data, and cruise point comprehensive constraint range data are generated; performing cruise node three-dimensional coordinate design on the three-dimensional coordinate data of the inspection point according to the cruise point comprehensive constraint range data to generate cruise point three-dimensional coordinate data; and performing optimized cruise path planning processing according to the cruise point three-dimensional coordinate data to generate optimized cruise path data. According to the invention, more accurate unmanned aerial vehicle tower space positioning and path planning are realized.
Owner:HUNAN ZHONGTUTONG UAV TECH CO LTD

AI workflow automatic management method based on creative scene

The invention provides an AI workflow automatic management method based on a creative scene, and relates to the technical field of data processing, and the method comprises the steps: obtaining initial data input by a user, extracting key elements from text data, analyzing related characteristics from non-text reference data, constructing an initial task chain, and analyzing a dependency relationship of task units, the method comprises the steps of generating a dependency path set, carrying out weight calculation on a dependency relationship among task units in the dependency path set, optimizing an initial task chain, generating a dynamic task chain table, selecting an applicable tool from an AI tool library, allocating a corresponding tool resource for each task unit, constructing an execution workflow, and dynamically running the execution workflow. Obtaining intermediate result data, detecting whether the intermediate result data meets the requirements of the logical relationship and the content style, and adjusting inconsistent parts to generate output result data; the intelligent degree of the workflow is improved.
Owner:XIAMEN ERWANLI CULTURE MEDIA CO LTD

Text semantic conflict detection method and system for constructing context association based on large model

The invention relates to a text semantic conflict detection method for constructing context association based on a large model. The method comprises the following steps: S1, segmenting an original text by using a segmentation function; s2, performing triple information extraction on each segmented text by using a large model to generate a data set of an entity corresponding relationship; s3, extracting data groups of all entity pairs and relationships, and performing confidence evaluation; s4, aggregating all triads, constructing a affair graph, and pointing out a contradictory relationship with relatively low confidence in the affair graph; and S5, for the found contradictory relationship, performing semantic conflict repair in the text in combination with the large model. According to the method, the accuracy of information generation of the large language model is remarkably improved in a context association mode.
Owner:SHANDONG ENERGY GRP CO LTD +1

Task execution method based on large language model and related equipment

The invention relates to the technical field of artificial intelligence, and provides a task execution method based on a large language model and related equipment. The method comprises the following steps: decomposing a complex task into a plurality of sub-tasks by utilizing a task decomposition model, and constructing an execution model for each sub-task; planning all the sub-tasks by using the task planning model to obtain an execution logic sequence of all the sub-tasks; calibrating each subtask to obtain a plurality of final subtasks, and then combining all the final subtasks according to an execution sequence to obtain a combined task; and executing the combined task by utilizing all the execution models to obtain a final execution result of the complex task. According to the method, the universality of the task execution method can be improved.
Owner:CENT SOUTH UNIV

Data development method and platform based on multi-source data integration

The invention belongs to the technical field of data processing, and discloses a data development method and platform based on multi-source data integration, and the method comprises the steps: collecting and preprocessing multi-source heterogeneous data, obtaining multi-source initial data, extracting initial features of different data sources, and forming a preliminary feature set; establishing and optimizing an event tree according to the collected multi-source heterogeneous data and the corresponding preliminary feature set, mining the multi-source data collected in real time based on the optimized event tree, analyzing the multi-source data collected in real time according to an established risk assessment model, performing comparative analysis on an actual risk event of an obtained risk prediction result, and obtaining a risk prediction result. Data access and operation are carried out through the user interface, efficient integration, processing, analysis and application of data are achieved, and the overall level of data development and utilization is improved.
Owner:江西省通信产业服务有限公司

Text structured extraction method and system, terminal and medium

The invention provides a text structured extraction method and system, a terminal and a medium, and particularly relates to the technical field of natural language processing and information retrieval, and the scheme comprises the following steps: obtaining an analysis text of a target text, adding an index number based on the analysis text, and splicing to obtain a target spliced text; obtaining a plurality of training data sets according to the plurality of texts; performing fine tuning on a preset large language model by using the plurality of training data sets to obtain a fine-tuned large language model; and based on the target spliced text, performing processing through a fine tuning large language model, and generating structured content of the target text. According to the scheme, the index numbers are fused into the text paragraphs, the key information of the target text can be pointed through the index numbers, the ability of a large language model to accurately and comprehensively understand the whole text structure is improved, it can be ensured that the structured content of the generated target text keeps the hierarchical structure of all the paragraphs in the text, and the user experience is improved. And the readability and accuracy of the structured content of the target text can be improved.
Owner:INT DIGITAL ECONOMY ACAD

Intelligent analysis system for initially created enterprise based on large language model

The invention provides a big language model-based intelligent analysis system for a newly created enterprise, and the system comprises a data collection module which is used for collecting data of the newly created enterprise; the data preprocessing module is used for preprocessing the data of the initially created enterprise; the vector database and knowledge graph construction module is used for constructing a vector database and a knowledge graph based on the preprocessed initially created enterprise data; the context information construction module is used for carrying out deep semantic analysis on related information, input by a user, of the to-be-analyzed enterprise, and fusing a semantic analysis result with data of the vector database and the knowledge graph to obtain context information; and the intelligent analysis module is used for carrying out intelligent analysis on the context information based on the initially created enterprise research analysis auxiliary large model, obtaining development state data of the to-be-analyzed enterprise, and completing initially created enterprise intelligent analysis based on the large language model. And comprehensive and efficient retrieval and processing of related data of the initially created enterprise are realized.
Owner:胡奇

Power industry knowledge question-answering method and system based on large language model

The invention provides a power industry knowledge question-answering method and system based on a large language model, and relates to the technical field of artificial intelligence, and the method comprises the steps: obtaining power industry data; constructing a knowledge graph with node weights; acquiring a training set taking the power industry question as a feature and taking the power industry answer as a label; constructing a generative question and answer model with knowledge graph embedded information; establishing a generation loss function; taking the generation loss function value smaller than a preset generation loss function value as a target, training the generative question-answer model to determine an optimal hyper-parameter of the generative question-answer model; obtaining a to-be-processed power industry problem; and inputting a to-be-processed power industry question into the trained generative question and answer model, and outputting an expected power industry answer corresponding to the to-be-processed power industry question. And in combination with the knowledge graph with the node weight and the generative question and answer model, the interference of training set noise on model output answers is reduced, and the consistency and accuracy of the generated answers are improved.
Owner:SHENZHEN YUANJIE MANAGEMENT CONSULTING CO LTD

Household electricity consumption analysis method based on non-intrusive load monitoring and related equipment

The invention belongs to the field of non-intrusive load monitoring, and discloses a household electricity consumption analysis method based on non-intrusive load monitoring and related equipment, firstly, non-intrusive monitoring is realized by using household total electricity meter data instead of a single equipment electricity meter, and the complexity and cost of household electricity consumption data collection are reduced; secondly, an unsupervised learning method is adopted to cluster the power load state data, data does not need to be marked in advance, and different load events are automatically identified and classified, so that the problems of load type diversity and high identification difficulty are effectively solved, and the accuracy and efficiency of load event detection are improved; in combination with a depth-first search algorithm and load event constraint conditions, fine search is carried out on load event labels, an undetermined event sequence set is generated, the logicality and continuity of event sequences are enhanced, electricity utilization behavior analysis of the electric equipment is deeper and finer, home users can better understand and optimize own electricity utilization modes, and the user experience is improved. And energy conservation, emission reduction and efficient utilization of energy are promoted.
Owner:STATE GRID BEIJING ELECTRIC POWER CO +1

Page component layout method and device, computer equipment and storage medium

The embodiment of the invention discloses a page component layout method and device, computer equipment and a storage medium. The method comprises the steps that a target layout resource and a target layout rule of a target low-code development platform are obtained, the target layout resource comprises a plurality of components carrying different element features, and the element features comprise size features, hierarchical relation features and element priority features; based on a grid layout algorithm, a target layout rule, the size features corresponding to the components and the hierarchical relation features, the multiple components are placed in a grid system of the current page, and a preliminary layout page of the target layout resources is obtained; based on a simulated annealing algorithm, a target layout rule, the hierarchical relation characteristics of the components and the element priority characteristics, performing position optimization on the components in the preliminary layout page to obtain a first target layout page; and outputting the first target layout page. By implementing the method provided by the embodiment of the invention, the layout of the middle page component of the low-code development platform can be automatically adjusted.
Owner:BEIJING LINGYI VISION TECH CO LTD

Geological text generation method, system and device based on large language model

The invention relates to the technical field of information processing, and particularly discloses a geological text generation method, system and device based on a large language model, and the method comprises the steps of text data collection, data set construction, data preprocessing, large model construction, model fine tuning and the like. And intelligent text generation of stratum lithology description based on the deep learning large language model is realized. According to the method, the ChatGLM big language model is trained, fine tuning is carried out based on the P-Tuning v2 technology, and meanwhile, the geotechnical field knowledge is fused, so that the model can understand and apply the professional knowledge of the geological lithology, the stratum lithology description text meeting the requirements of the engineering geological segmentation specification is intelligently generated according to the known lithology description information, the continuity is good, the logicality is high, and the method is suitable for being used in the field of engineering geology. The whole method is simple to operate, can efficiently improve the efficiency of manual data arrangement and analysis, and is convenient to popularize and apply.
Owner:CHINA RAILWAY ERYUAN ENGINEERING GROUP CO LTD +1

Intelligent answering method, system and device for complex reasoning problem and medium

ActiveCN119067223AEnhance multi-step reasoning capabilitiesincrease flexibilityInference methodsPersonalizationQuestion analysis
The invention discloses an intelligent answering method, system and device for complex reasoning questions and a medium, which are in one-to-one correspondence schemes, and in the scheme, a coarse granularity-fine granularity reasoning normal form is provided, and a final reasoning tree is constructed through circulation of three stages: a decomposition stage: decomposing a complex question into a plurality of simple sub-questions, each sub-problem can be understood and processed in an analysis stage, each sub-problem is deeply analyzed, a complete reasoning process is established, the correctness and logicality of the sub-problems are evaluated and checked, in a reflection stage, the whole problem solving process is reviewed, errors or defects possibly existing in the problem solving process are analyzed, and targeted correction is provided. Through the cycle process, the method not only can efficiently and accurately model a complex problem reasoning process, but also can improve the accuracy and efficiency of reasoning question and answer related tasks, can provide a high-interpretability problem solving logic thought for personalized online learning guidance, and has relatively high educational and practical values.
Owner:UNIV OF SCI & TECH OF CHINA

Voice separation method, device and equipment, storage medium and voice transfer method

The invention relates to a voice separation method and device, computer equipment, a storage medium and a voice transfer method. The method comprises the following steps: acquiring multi-channel mixed voice data; separating single-path voice data from the multi-channel mixed voice data; determining identity conversion time points of different speakers from the single-path voice data, and segmenting the single-path voice data into a plurality of voice short segments of different speakers according to the time points; classifying the voice short segments of a plurality of different speakers to obtain voice short segments belonging to the same speaker; and carrying out speaker clustering processing on the voice short segments of the same speaker to obtain separated voice data belonging to the same speaker. By adopting the method, the accuracy of voice separation can be improved, and the accuracy of voice transfer is further improved.
Owner:VOLKSWAGEN-MOBVOI (BEIJING) INFORMATION TECH CO LTD

Question and answer method, device and equipment based on large language model

The invention provides a question answering method, device and equipment based on a large language model, and the method comprises the steps: determining a corresponding answering process based on the obtained question content, and determining a plurality of keywords and questions corresponding to a current stage; inputting the keyword into a keyword retrieval model to obtain a first retrieval result, inputting the question into a vector retrieval model to obtain a second retrieval result, and inputting the first retrieval result and the second retrieval result into a fine arrangement model to obtain a corresponding reference document; based on the question content, the answer process and the reference document, determining corresponding answers, and summarizing the answers to the generated answers; judging whether a subsequent stage in the reflection answering process needs to be adjusted or not, if so, adjusting, repeating the steps, and determining an answer corresponding to the next stage; and traversing each stage to obtain a corresponding answer, and taking the generated answer as a final answer corresponding to the question content. According to the method, the illusion problem of the LLM can be reduced, and the credibility of the final answer is improved.
Owner:上海蜜度蜜巢智能科技有限公司

Automatic production line for intelligent household appliance touch glass panels

The invention relates to the technical field of glass panel production, in particular to an intelligent household appliance touch glass panel automatic production line which comprises glass melting equipment, environmental protection equipment, glass cutting equipment, glass recycling equipment and glass toughening equipment. The glass recycling equipment comprises a rack, a glass conveying device and a glass recycling device, wherein a diamond file is arranged on the rack; the adsorption assembly is arranged on the rack, and a trigger plate is arranged on the adsorption assembly; the power assembly is connected with the diamond file, a pulley is installed on the power assembly, and the pulley is matched with the trigger plate so that the adsorption assembly can generate negative pressure; the elastic supporting assembly is arranged between the diamond file and the power assembly, and the elastic supporting assembly is matched with a side plate arranged on the rack, so that the diamond file has two stations, namely a separation station and a scribing station; after the adsorption assembly generates negative pressure, the diamond file is switched to the scribing station so as to scribe the glass before crushing, so that the glass is crushed step by step during crushing, and splashing of glass residues is avoided.
Owner:SUZHOU LINGTONG SAFETY GLASS CO LTD

Vehicle energy consumption monitoring method and system based on multi-source data

ActiveCN118859813BAccurately locate relative positionspositioning relative positionProgramme controlComputer controlData setMulti source data
The present invention discloses a vehicle energy consumption monitoring method and system based on multi-source data, which relates to the field of vehicle energy consumption monitoring. The vehicle energy consumption monitoring method based on multi-source data includes: obtaining initial vehicle measurement data sent by a sensor module installed on a vehicle, and optimizing the initial vehicle measurement data using a time window filtering technique to obtain a vehicle internal measurement data set; identifying energy consumption influencing factors of the vehicle's external environment and generating a vehicle energy consumption external influence sequence; and generating a number of vehicle energy consumption monitoring scenarios based on the vehicle internal measurement data set and the vehicle energy consumption external influence sequence. The present invention combines the vehicle internal measurement data and the external influencing factor sequence, and uses neighborhood mutual information and autoencoder technology for feature fusion to construct an energy consumption monitoring scenario that accurately matches actual driving conditions, so that energy consumption anomalies can be quickly identified and responded to between associated scenarios.
Owner:ANHUI HEFEITER AUTOMOBILE CO LTD

Large transport vehicle identification method and system based on radar and vision fusion

The invention relates to the technical field of data processing, in particular to a large transport vehicle identification method and system based on radar and vision fusion. The method comprises the following steps: collecting a monitoring radar signal on the way of a transport vehicle through radar monitoring equipment; carrying out transport vehicle characteristic analysis on the radar signals according to the transport vehicle on-the-way monitoring radar signals, and generating transport vehicle characteristic data; collecting on-way visual data of the transport vehicle by using high-resolution monitoring equipment; carrying out transport vehicle on-way visual feature analysis according to the transport vehicle on-way visual data, and generating transport vehicle on-way visual feature data; and according to the transport vehicle characteristic data and the transport vehicle on-way visual characteristic data, carrying out transport vehicle external heterogeneous data mapping and transport vehicle dynamic characteristic mode identification, and respectively generating a transport vehicle external heterogeneous data matrix and transport vehicle dynamic characteristic mode data. According to the method, the vehicle identification of the large transport vehicle is carried out based on radar and vision fusion, and the vehicle information of the large transport vehicle is identified more comprehensively.
Owner:HUNAN COMM RES INST CO LTD +1

Abstract text generation method and device, equipment and medium

The invention belongs to the field of artificial intelligence, and relates to an abstract text generation method which comprises the following steps: acquiring a to-be-processed text, and encoding the to-be-processed text to obtain a plurality of segmented words of the to-be-processed text and encoding information corresponding to each segmented word; determining key segmented words from the plurality of segmented words, and constructing an adjacent matrix of the to-be-processed text based on the key segmented words; performing feature fusion on the coding information and the adjacent matrix to obtain a node feature matrix of the to-be-processed text; performing feature optimization processing on the node feature matrix to obtain a target node feature matrix of the to-be-processed text; and generating an abstract text of the to-be-processed text based on the target node feature matrix. The invention further provides a device, equipment and a medium. In addition, the invention also relates to a block chain technology, and the to-be-processed text, the plurality of segmented words, the coding information, the adjacency matrix, the node feature matrix, the target node feature matrix and the abstract text can be stored in the block chain. The abstract text generation accuracy can be improved.
Owner:PING AN TECH (SHENZHEN) CO LTD

Robot tracking control learning method based on postmortem screening experience playback

The invention relates to the field of mechanical arm trajectory tracking, in particular to a robot tracking control learning method based on postmortem screening experience playback, which comprises the following steps: initializing target network parameters and an experience pool, and storing a state transition tuple in the experience pool; when the number of the state transition tuples in the experience pool is larger than the preset number of the state transition tuples, the action complex state is determined according to the track smoothness and the node covariance degree of the robot mechanical arm; determining an experience pool optimization mode according to the complex state of the action; selecting experience of which the estimated position deviation is greater than or equal to the standard estimated position deviation as screening experience, and determining an experience selection mode according to the number of the screening experience; updating the cr-it ic network and the actor network, and carrying out soft updating on the target network of the cr-it ic network and the target network of the actor network respectively; and according to the reward evaluation value, soft update weights of the target network of the cr-it-ic network and the target network of the actor network are adjusted. According to the method, the tracking effect and the tracking stability of the model are improved.
Owner:DONGGUAN UNIV OF TECH

A modal balance and semantic enhancement training method for a full-modal safety fence

PendingCN122594864AComprehensive security coveragefull coverage status
The application discloses a modal balance and semantic enhancement training method of a full-modal safety guardrail, which significantly reduces the missed detection and false rejection of cross-modal queries by improving the training data and designing the training target, and improves the safety and reliability of the full-modal guardrail. The application efficiently constructs modal de-biased data, constructs a multi-modal query quadruple based on semantic pairing of a text seed query, and enumerates combinations of modal subsets and safety semantic labels to construct risk variant query samples. The data construction method can efficiently expand the training data based on a small amount of text samples, so that the safety and unsafe states of various modal combinations are comprehensively covered, the model is prevented from learning modal shortcuts during the training process, the dependence of the guardrail on different modal safety semantics is rebalanced, fine-grained safety supervision signals are efficiently expanded, and the cross-modal safety semantic understanding of the guardrail model is improved.
Owner:ZHEJIANG UNIV

Data generation method and device, electronic equipment and storage medium

The embodiment of the invention discloses a data generation method and device, electronic equipment and a storage medium, and relates to the technical field of computer vision and model generation, and the method comprises the steps: carrying out the training of an image segmentation large model through an image data set, obtaining a trained segmentation model, obtaining a to-be-processed image, and obtaining a to-be-processed image; performing semantic segmentation on the to-be-processed image through the segmentation model to obtain a category prediction map of each pixel point in the to-be-processed image, performing semantic recognition according to the category prediction map of each pixel point to obtain a plurality of semantic regions of the to-be-processed image, adding a corresponding enhancement model MOD for each semantic region, and performing overlapping, position and logic judgment to obtain a plurality of semantic regions of the to-be-processed image. And adjusting the enhancement model MOD according to a judgment result to generate image data. By integrating image segmentation, region identification and MOD addition, the problems of high technical threshold, insufficient authenticity and diversity of generated data, target overlapping, position error and logic disorder in the prior art are solved.
Owner:SHENZHEN TIANHAI CHENGUANG TECH CO LTD

Power load characteristic multi-scale analysis method based on multi-source data and dynamic clustering

The invention discloses a power load characteristic multi-scale analysis method based on multi-source data and dynamic clustering, and relates to the technical field of power analysis, and the method comprises the following steps: fusing power, production, policy, environment and other multi-source heterogeneous data, and carrying out space-time alignment; cleaning the load data and identifying a characteristic time period influenced by the specific factor according to the auxiliary information; decomposing the load curve and recombining the load curve into components with definite physical significance according to auxiliary data such as a production time sequence; performing context-fused dynamic clustering and mode correlation analysis on each load component according to the time scale; and extracting refined characteristics and adjustable potential indexes, and generating a visual resource map capable of supporting scheduling. According to the method, by introducing a polymorphic load component separation and multi-scale collaborative clustering mechanism, cause decoupling and refined mode recognition of complex industrial loads can be realized, and the problems that in an existing method, mode confusion is likely to occur, and load characteristics under different driving mechanisms are difficult to distinguish are solved.
Owner:STATE GRID JIANGSU ELECTRIC POWER CO LTD TAIZHOU POWER SUPPLY BRANCH

A distributed critical path mining method for billion-level enterprise transaction networks

The present invention is a distributed key path mining method for a billion-level enterprise transaction network, and belongs to the field of computer technology. In view of the difficulty in efficiently identifying key transaction paths under large-scale nodes at present, and the poor interpretability of the identified paths, the method of the present invention first splits the enterprise transaction network into a subgraph centered on each enterprise node, loads the subgraph into Pandas UDF, uses GAT to calculate the attention weight of the central node to the first-order neighbors, and aggregates the transaction information of the second-order neighbors, and then selects the first ten neighbors with large attention weights for attention-guided random walks; the path importance score is calculated based on the attention weight for the path obtained by the walk, and the key transaction path and key nodes are identified based on this. The present invention can capture and quantify the transaction influence between enterprise nodes, guide the random walk process to focus more on the key transaction path, improve the accuracy of path selection, and enhance the ability to explain important transaction relationships between enterprises when the path is obtained.
Owner:BEIJING DAOKOU JINKE TECH CO LTD

Intelligent recommendation system and method for semiconductor thin film deposition process based on knowledge graph

PendingCN122264096ARealize intelligent recommendationEfficient Decision SupportForecastingInference methodsAlgorithmTheoretical computer science
The application discloses a semiconductor thin film deposition process intelligent recommendation system and method based on a knowledge graph, belongs to the technical field of semiconductor manufacturing processes, and comprises the following: a semiconductor thin film deposition process knowledge graph construction module; a prompt word template generation module: used for receiving process requirements and constraint conditions input by a user, performing semantic analysis on the process requirements and constraint conditions, filling the analysis results into a preset prompt word template, and outputting standardized instructions meeting the calling requirements of a large language model; and a large language model reasoning module: used for receiving the standardized instructions, performing semantic retrieval and logical reasoning on the standardized instructions within the constraint range of the semiconductor thin film deposition process knowledge graph, generating a plurality of candidate thin film deposition process schemes meeting process constraint conditions based on entity relationships and causal associations in the semiconductor thin film deposition process knowledge graph; and the application constructs a complete process recommendation solution and exhibits significant advantages in multiple aspects.
Owner:INTELLIGENT COMPUTING MATRIX (SHENZHEN) DATA TECHNOLOGY CO LTD

Building energy consumption analysis method and system based on MCP service model and agent

ActiveCN121809861AImplement automatic schedulingAchieve fully automatic generationArtificial lifeResourcesService modelLinguistic model
The invention provides a building energy consumption analysis method and system based on an MCP service model and an intelligent agent, and relates to the technical field of building energy saving and artificial intelligence, and the method comprises the steps: constructing and packaging an MCP service model, and abstracting the MCP service model into a standardized function interface after packaging; acquiring and storing operation equipment and equipment operation parameters in the building; an input request of a user is obtained, a task intention in the input request is recognized, the task intention is converted into specific API calling according to the task intention, and the API obtains structured energy consumption data by calling an MCP function interface; combining the structured energy consumption data returned by the MCP with the cue word template, transmitting the combined Agents to the large language model, and outputting to obtain a report; and visualizing or exporting the report. According to the invention, full-automatic generation from complex building energy consumption data to a structured report containing natural language analysis and visual charts is realized.
Owner:SHANDONG UNIV

Operating system testing method, device and system integrating multiple chips

ActiveCN118897810BImplement integration testsshorten the timeComputer hardwareTest efficiency
The present application discloses a method, device and system for testing an operating system integrating multiple chips, belonging to the field of chip testing technology. The method includes: receiving input data from a user in a running panel of a test script; the test script is used to test multiple different chips with at least partially identical functional instructions; selecting a target chip for testing according to the input data; using the test script to test the operating system of the target chip using a target command parameter set. The present application establishes the same test script for multiple different chips with at least partially identical functional instructions, tests different chips by changing input data and command parameters, and realizes integrated testing of multiple chips with a single test process. Different chips can also interact with each other, and different command parameters can be tested by modifying the value of the command parameter. There is no need to write code for each command parameter test, which reduces code redundancy and improves test efficiency.
Owner:BEIJING SMARTCHIP MICROELECTRONICS TECHNOLOGY CO LTD

An accident scene screening matching device and method for SOTIF test requirements

This invention relates to the field of expected functional safety testing technology for autonomous vehicles, specifically to an accident scenario screening and matching device and method for SOTIF testing requirements. The device includes a processing system comprising an acquisition module, an analysis module, a risk modeling module, and a matching module. The risk modeling module uses Fault Tree Analysis (FTA) to perform a structured decomposition of SOTIF risk causative factors, quantifies the importance of risk elements through a combination of subjective and objective weighting, and outputs a comprehensive weight for each risk element. The matching module combines the comprehensive weight of risk elements to weight and enhance the accident scenario text, filtering out accident scenarios that match the test requirement text. This invention achieves precise matching between test requirements and accident scenarios by semantically associating risk elements with accident scenario text in an accident database.
Owner:XIHUA UNIV

Government affair knowledge graph construction method and system based on artificial intelligence

The invention relates to the technical field of government affair informatization, in particular to a government affair knowledge graph construction method and system based on artificial intelligence, and the method comprises the steps: obtaining multiple pieces of multi-source government affair data, carrying out the format unification of the multiple pieces of multi-source government affair data, obtaining multiple unified texts, carrying out the cleaning segmentation of the multiple unified texts, obtaining multiple text segments, and carrying out the segmentation of the multiple text segments; performing subject recognition on the plurality of text segments to obtain a plurality of government affair entities, performing normalized alignment on the plurality of government affair entities to obtain a plurality of unique entities, obtaining a plurality of semantic relationships of the plurality of unique entities, and constructing a triple set by using the plurality of unique entities and the plurality of semantic relationships, and constructing a government affair knowledge graph by utilizing a plurality of triads in the triad set, and automatically generating a rule required by the rule engine by the government affair knowledge graph. The method can solve the problem that intelligent application of government affair knowledge is difficult due to dispersion of multi-source government affair data and low manual integration efficiency.
Owner:SHENZHEN HUAYUNJIA INTELLIGENT TECHNOLOGY CO LTD