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

53 results about "Relational structure" patented technology

Relational structure. (data structure) Definition: The counterpart in formal logic of a data structure or class instance in the object-oriented sense. Examples are strings, directed graphs, and undirected graphs. Sets of relational structures generalize the notion of languages as sets of strings.

Document-level relation extraction method based on mixed prompt tuning

The invention provides a document-level relation extraction method based on mixed prompt tuning, which comprises the following steps of: constructing a document-level knowledge graph module by utilizing historical documents, and acquiring and processing entity and relation structure information; a sentence-level semantic retrieval enhancement mechanism is designed, and context information is matched and retrieved based on semantic relevancy; providing a mixed prompt tuning method, and performing efficient parameter fine tuning on the large language model; and finally, aggregating the document-level knowledge graph and sentence-level semantic retrieval to enhance and generate an output result of a SetRAG mechanism, and carrying out document-level relationship prediction. According to the method, the domain knowledge based on the atlas and the semantic knowledge based on retrieval are combined, so that the problems that entities and relation interaction information are difficult to capture and cross-sentence relation reasoning is difficult in existing document-level relation extraction can be effectively solved, and meanwhile, efficient and accurate document-level relation extraction is realized; and the performance of the public test data set is obviously improved.
Owner:NANJING UNIV OF INFORMATION SCI & TECH

File information association analysis method based on heterogeneous knowledge graph

The invention discloses a file information association analysis method based on a heterogeneous knowledge graph, which comprises the following steps: collecting and preprocessing multi-modal heterogeneous file information to generate a standardized input data set; establishing a heterogeneous knowledge graph based on the standardized data, and constructing a file and relation structure; modeling a file semantic space by adopting an improved TransH model, and generating semantic embedding representation; a semantic-time double-flow mechanism is introduced, and semantic flow and time flow joint representation is established; a multi-agent collaborative distillation system is constructed, and semantic, time and structure fusion is achieved; and calculating a comprehensive association score of the file and triggering incremental updating and synchronous correction. According to the method, the heterogeneous knowledge graph is constructed, and the improved TransH model, the semantic-time double-fluid mechanism and the multi-agent collaborative distillation system are combined, so that dynamic semantic modeling and accurate association analysis of multi-modal file information are realized.
Owner:BEIJING POLYTECHNIC

Multi-source heterogeneous data entity alignment method and device and medium

The invention provides a multi-source heterogeneous data entity alignment method and device and a medium, and relates to the technical field of artificial intelligence. The method comprises the following steps: merging two knowledge maps according to a predicate alignment algorithm to obtain a merged knowledge map; learning semantic abstract embedding and relation structure embedding of each entity in the combined knowledge graph; according to the dependency relationship of words in the attribute text of each entity, performing part-of-speech learning on attribute semantic embedding of each entity; determining a combined embedding vector of each entity according to semantic abstract embedding, relation structure embedding and attribute semantic embedding of each entity; and aligning the entities in the combined knowledge graph according to the combined embedding vectors of the entities. According to the method, the accuracy of knowledge graph entity alignment in the scene of insufficient seed quantity and sparse attribute is effectively improved.
Owner:INST OF COMPUTING TECH CHINA ACAD OF RAILWAY SCI +2

Ship knowledge graph construction method based on large model and graph neural network

The invention discloses a ship knowledge graph construction method based on a large model and a graph neural network, and the method comprises the steps: carrying out the format conversion and protection type partitioning processing of a professional document, and completing the recursive segmentation through a placeholder protection formula, a table and other structures according to the semantic hierarchy, and obtaining a text block suitable for the extraction of an entity and a relation; extracting domain entities by utilizing a large language model containing thinking chain cues, and positioning candidate entities in combination with an AC automaton so as to improve the coverage rate and accuracy of relation triple extraction; performing new entity backfilling and relation standardization on an extraction result, and constructing an initial knowledge graph with a consistent structure; and inputting the initial knowledge graph into a graph neural network model with directional expansion and a multi-scale decoder, and complementing a missing relationship through link prediction, so that node distribution and a relationship structure of the graph are more complete. According to the method, a continuous processing chain from text preprocessing, knowledge extraction to graph completion is formed.
Owner:SHANGHAI JIAOTONG UNIV

Method and system for automatically generating and updating multi-split system diagram based on binary tree

The invention discloses a method and a system for automatically generating and updating a multi-split system diagram based on a binary tree. The method comprises the following steps: setting the style of the multi-split system diagram and the type of associated labeling parameters; reading floor information of the BIM model, traversing floors, and reading an outdoor unit of each floor; each outdoor unit is used as a root node of a binary tree, other connected devices are searched as child nodes, a binary tree type relation structure is constructed, and a binary tree array is generated; a floor table and a drawing name are generated, the binary tree array is traversed, a preorder traversal mode is adopted for each binary tree, a corresponding equipment graph is drawn on a corresponding floor in the floor table, marking is conducted, and a multi-split system diagram is obtained; if the BIM model is modified, the binary tree type relation structure is updated according to modification in the BIM model, and the updated multi-split system diagram is regenerated. The problems that the multi-split system diagram is difficult to automatically generate and change, low in efficiency and prone to errors are solved.
Owner:CHINA MACHINERY INT ENG DESIGN & RES INST

Multi-modal patent semantic retrieval method and system

The invention belongs to the technical field of information retrieval, and discloses a multi-modal patent semantic retrieval method and system.The reliable cross-modal relation is recognized by establishing relation anchor points, extracting local cross-modal relation representation around the anchor points and verifying the cross-modal consistency of the relations, and further, the multi-modal patent semantic retrieval method and system have the advantages that the reliability of the multi-modal patent semantic retrieval is improved; technical features / components and successfully aligned cross-modal relationships are constructed into graph structure representation, and similarity calculation is performed by using a graph matching algorithm, so that the overall logic and details of technical schemes can be more effectively understood, schemes only containing the same technical features but having different relationship structures can be distinguished, and the accuracy of the technical schemes can be improved. The accuracy and correlation of multi-modal patent retrieval can be improved.
Owner:SHENZHEN MUNICIPAL CUBE TECHNOLOGY CO LTD

Abnormal risk prediction method and device based on assembly knowledge graph, equipment and medium

The invention discloses an abnormal risk prediction method and device based on an assembly knowledge graph, equipment and a medium, and relates to the technical field of data processing. The method comprises the steps of obtaining multi-source heterogeneous data, performing association and alignment through a unified primary key set, and performing segmentation to generate a time slice data set; and constructing an assembly domain ontology and an assembly knowledge graph mode layer, instantiating the assembly domain ontology by using the time slice data set, and constructing an assembly knowledge graph snapshot sequence. And based on the assembly knowledge graph snapshot sequence, modeling and updating the influence relationship between variables in the assembly process to obtain an assembly process influence relationship structure. And in combination with the assembly knowledge graph snapshot sequence and the assembly process influence relation structure, coding is carried out, the state representation of the next time window is predicted based on the embedding representation of the historical time window, and an abnormal risk score is calculated. When the risk exceeds a threshold value, influence path tracking is carried out, controllable decision variables are screened, candidate intervention schemes are generated, the effects of the candidate intervention schemes are evaluated, and decision suggestions are output.
Owner:XIAMEN UNIV OF TECH

Zero-sample composite fault diagnosis method based on semantic graph embedding and multi-stage fusion

The invention discloses a zero-sample composite fault diagnosis method based on semantic graph embedding and multi-stage fusion, and the method comprises the following steps: firstly, generating an initial semantic descriptor through manual definition or statistical features according to known fault types, and constructing a fault relation structure diagram to represent the association between types; combining reconstruction, semantic comparison and propagation loss by using a graph convolutional network, and fusing the initial semantics and the relation graph to generate enhanced semantic features; meanwhile, a multi-modal model is constructed to extract vibration, temperature, acoustics and other signal features, and after multi-stage fusion of input-stage cross-modal attention, feature-stage Transform and output-stage semantic alignment, a semantic feature supervision training network is jointly enhanced; and finally, extracting unseen fault features in a zero sample scene and carrying out classified diagnosis. According to the method, through a multi-modal fusion and semantic enhancement strategy, the precision and generalization ability of zero-sample composite fault diagnosis are remarkably improved.
Owner:YANGTZE DELTA REGION INST OF UNIV OF ELECTRONICS SCI & TECH OF CHINE (HUZHOU)

Chip IP interface information automatic analysis and connection modeling method

The invention discloses a chip IP interface information automatic analysis and connection modeling method, and relates to the technical field of integrated circuit verification, and the method comprises the steps: multi-source interface information collection: docking a design server through an FTP / SFTP protocol, automatically retrieving and capturing a target file, carrying out format preprocessing, and storing the collected original information through a database; heterogeneous information fusion analysis: extracting port physical attributes and generating machine executable rules by constructing a grammar analyzer and a semantic analyzer respectively, and automatically judging multi-source information contradictions by adopting a three-level judgment mechanism; structuring modeling of the connection relation: constructing a directed graph model of the IP nodes, the signal nodes and the constraint nodes, and automatically executing directed graph model verification; and through a standardized API interface docking verification tool, outputting a model file and a report file in standardized JSON-LD and DOT formats. Through cooperative work of cross-format acquisition, double-engine analysis, structured modeling and standardized output, full-automatic processing of IP interface information is realized.
Owner:JIANGSU XINSHENG INTELLIGENT TECH CO LTD

Method, system and device for automatically constructing industrial knowledge graph and storage medium

The invention discloses a method, system and device for automatically constructing an industrial knowledge graph and a storage medium. The method comprises the steps that an embedded layer is established for unstructured data, words are mapped into word vectors, entities and categories in the word vectors are obtained through a feedforward neural network, and a first knowledge graph is constructed after the relation between the entities is judged and extracted; mapping the data structure into an ontology and a relationship by using a triple for the structured data, and constructing a second knowledge graph after obtaining a relationship structure between entities according to the mapping relationship; and periodically executing a data capture task, detecting updated data, performing dynamic knowledge fusion, and updating and optimizing the first knowledge graph and the second knowledge graph. According to the method, the conversion from the structured data to the ternary relationship can be realized, and a large amount of manual annotation and knowledge graph generation processes are reduced; and customized construction of the enterprise-level data knowledge graph can be realized, and data mining analysis is facilitated, so that combination of multiple systems and integration and sharing of data are realized.
Owner:LUCULENT SMART TECHNOLOGIES CO LTD

Power grid equipment fault prediction and maintenance scheme generation method

The invention provides a power grid equipment fault prediction and maintenance scheme generation method. The method comprises the following steps: acquiring a data set; standardizing the data set to obtain a structured data set; based on the structured data set, defining a quintuple relation structure, and establishing a dynamic knowledge graph; constructing an encoder group; based on the data set and the dynamic knowledge graph, fusing the time sequence feature, the text feature and the graph feature by using an encoder group to obtain a fused feature; based on the fusion feature, obtaining a fault classification probability and a decomposition cost item; according to the fault classification probability and the decomposition cost item, determining a corresponding measure set and a feasible time window set in the dynamic knowledge graph to obtain a search space; and based on the search space, determining a target maintenance scheme so as to solve the problem that when an operation maintenance scheme is formulated by a power grid enterprise at present, systematic consideration on the equipment fault probability, the disposal cost and the resource configuration is often lacked, so that the scheme selection cost is high or execution conflicts exist.
Owner:DEZHOU POWER SUPPLY COMPANY OF STATE GRID SHANDONG ELECTRIC POWER +2

Micro-video popularity prediction method based on graph prompt enhancement

The invention discloses a micro-video popularity prediction method based on graph prompt enhancement, relates to the technical field of machine learning, and mainly utilizes a generation retriever driven by a large-scale multi-modal model to create retrieval query with rich context and execute retrieval so as to form a dynamic micro-video support set which effectively reflects an evolution mode. On this basis, the graph-based prompter generates evolutionary prompts by capturing time and relational structures in the support set. The prompts represent the latest trend dynamics, and can be used as a few examples for guiding the pre-training visual language model. By integrating the evolution prompts, the PVLs can more accurately simulate the constantly changing popularity trend, so that stronger and more predictive representation is generated, and the purpose of enhancing popularity prediction is achieved.
Owner:郑州埃文科技有限公司

Code generation method and corresponding device

The invention provides a code generation method which comprises the steps that a first index used for accessing a knowledge base is determined according to a target problem, the target problem is a function to be achieved through a target code described through a natural language, the knowledge base comprises a mapping relation between a code segment and a structured view, and the structured view is used for representing a semantic and logical relation of the code segment; according to the first index, querying a knowledge base to obtain a first mapping relation associated with the first index; and inputting the target question and prompt information into a code generation model (LLM) to obtain a target code of the target question, the prompt information being associated with the first mapping relationship. According to the scheme provided by the invention, when the code is generated, the target code is generated in combination with the prompt information, and the prompt information can stimulate the context learning ability of LLM, so that the code generation effect is improved, and the accuracy of the generated target code is improved.
Owner:HUAWEI TECH CO LTD

A method for threat intelligence aggregation and attack anomaly feature intelligent generation based on large language model

The present invention discloses a method for intelligently generating threat intelligence aggregation and attack anomaly features based on a large language model, including obtaining network security intelligence of an open source threat intelligence community platform in real time; optimizing the prompt words of the large language model according to predefined multidimensional feature attributes, and extracting multidimensional feature attributes of threat intelligence from network security intelligence; constructing a mesh graph relationship structure database of threat intelligence according to the extracted multidimensional feature attributes of threat intelligence; constructing a threat intelligence knowledge graph according to the mesh graph relationship structure database; querying the threat intelligence knowledge graph, marking attack behavior features, connecting attack behavior feature points, and generating an attack behavior feature subgraph. The method of the present invention enables the security team to quickly respond to emerging threats and improves the ability to respond to network security threats.
Owner:JIANGSU ELECTRIC POWER INFORMATION TECH

Data model establishment method and device, and clinical auxiliary decision method and device

The present disclosure relates to a data model establishing method and device, and a clinical auxiliary decision method and device. The data model establishing method comprises: field disassembling historical data according to different disease categories; structured processing is performed according to the fields obtained by disassembling to form a structured output rule; and a disease data model is established according to the structured output rule combined with clinical data. The present disclosure also relates to a data model establishing device, an electronic device and a computer readable medium, and a clinical auxiliary decision method, device, electronic device and computer readable medium based on the data model establishing method. The structured output rule is generated by structured processing of diseases, and the clinical data is saved in a relational structure in a database according to the rule to form a data model, so that the clinical data can be effectively utilized, the interconnection and intercommunication between data can be realized, and the research conversion rate can be improved.
Owner:TIANJIN HAPPY LIFE TECH CO LTD +1

Method and system for automatic generation and update of multi-connection system graph based on binary tree

The application discloses a kind of based on binary tree's multi-connection system diagram automatic generation and updating method and system, comprising: setting the style of multi-connection system diagram, the type of associated label parameter;Read the floor information of BIM model, traverse floor, read the outdoor unit of each floor;With each outdoor unit as the root node of binary tree, find the other device connected as child node, construct binary tree type relationship structure, generate binary tree array;Generate floor table and drawing name, traverse the binary tree array, on the corresponding floor in floor table corresponding device graphics is drawn using the mode of pre-order traversal to each binary tree, and is labeled, obtains multi-connection system diagram;If BIM model is modified, according to the modification in BIM model updates binary tree type relationship structure, and generates the multi-connection system diagram after being updated again.The application solves the problem that multi-connection system diagram is automatically generated difficult, changes difficult, low efficiency and error-prone.
Owner:CHINA MACHINERY INT ENG DESIGN & RES INST

A cross-language entity alignment method and system based on relationship semantic enhancement

The application relates to the field of natural language processing, and discloses a cross-language entity alignment method and system based on relationship semantic enhancement, which comprises the following steps: constructing entity initial embedding and relationship embedding representation of a cross-language knowledge graph; a key neighbor screening stage; screening key neighbor entities with structural matching and semantic matching based on an entity similarity matrix and a relationship similarity matrix and in combination with a preset semantic correlation threshold; neighbor-enhanced entity embedding updating; adopting a graph attention network to fuse semantic features of target entities and key neighbors thereof, and generating context-aware entity embedding; semantic consistency alignment propagation: first-order entity feature alignment and second-order relationship structure alignment are jointly optimized through a shared permutation matrix; dynamic seed expansion iteration: high-confidence unaligned entity pairs are dynamically identified according to Manhattan distances of entity embedding, entity pairs with distances lower than a threshold are added to a seed set, and the foregoing steps are iteratively executed until convergence.
Owner:BEIJING PEOPLE'S POLICE COLLEGE

DAG-based block chain consensus method

The invention discloses a DAG-based block chain consensus method, which ensures the uniqueness and availability of blocks in a DAG by constructing a reliable broadcast algorithm under a new hybrid model. And meanwhile, throughput is improved by utilizing a DAG structure, a rapid confirmation path is fused and added to realize rapid confirmation and reduce delay, and then the activity of the protocol is ensured through a coin confirmation path. According to the method, a main-free design is introduced into a hybrid algorithm design, a multi-proposal architecture is adopted, block relationships of blocks of proposals are structured through a directed acyclic graph (DAG), and the activity and security of a system are ensured by using a random algorithm and a time hypothesis. In addition, by introducing a rapid confirmation condition, the delay is reduced while the throughput capacity of the protocol is ensured. The consensus of the blocks can be completed in the fastest four rounds of communication, and most of the blocks can complete the consensus of the blocks in the six rounds of communication, so that the low time delay of the block consensus process is ensured, and the responsiveness of the system is improved.
Owner:SHANGHAI JIAOTONG UNIV

A Judicial Information Retrieval Method and System Based on Graph Neural Networks

This invention relates to the field of data processing technology, and in particular to a judicial information retrieval method and system based on graph neural networks. The method includes: extracting key information from judicial case text data using named entity recognition and relation extraction tools to obtain legal fact case subgraphs and legal issue case subgraphs; enhancing the structural information of the legal fact case subgraphs and legal issue case subgraphs through topology enhancement; encoding case text information and structural information using a pre-trained language model; guiding model training based on a multi-level loss function system, integrating cross-subgraph contrast loss and structural consistency loss; and using a graph neural network model to introduce a global attention mechanism for feature extraction and aggregation of the legal fact case subgraphs and legal issue case subgraphs. This model effectively encodes the complex relational structures in legal documents using graph learning algorithms, language modeling techniques, and contrastive learning methods, thereby achieving accurate matching and retrieval of relevant cases.
Owner:中国司法大数据研究院有限公司

Computer implemented method for creating database structures without knowledge on functioning of relational database system

Computer implemented methods for storing ad hoc relations between previously unrelated database objects assigned to different database structures using an electronic computing device are presented, the methods including: causing the electronic computing device to define at least three database structures: structure A, structure B and structure C, where each of the at least three database structures each includes a number of objects, where there exists at least one relation between objects of structure A and objects of structure B, and where there exists at least one relation between objects of structure B and objects of structure C; filtering data of structure A; accessing structure B using a first selected relation between structure A and structure B; storing information about filtering of structure A and information on a first selected path between structure A and structure B; filtering results obtained from structure B.
Owner:PILAB

A low-code platform data modeling method and system based on Dataset technology

This application provides a low-code platform data modeling method and system based on Dataset technology, relating to the field of low-code technology. The method acquires configuration information of a target business object, generates structured metadata based on the configuration information, and generates an application component including a strategy evaluation engine. During the operation of the application component, the engine monitors the data status in real time and dynamically adjusts the behavior or relational structure of the application component when the policy rule conditions are met, without modifying the generated source code. To improve the efficiency of large-scale rule processing, the method extracts conditional expressions and adjustment instructions from the policy rules, identifies dependencies, calculates structural and temporal correlation weights accordingly, generates predictive hybrid weights, and performs clustering. This solution improves strategy evaluation performance, supports dynamic business adaptive modeling, and is suitable for digital transformation scenarios.
Owner:JIANGSU JINNONG

Language conversion model training method, conversion method, device, apparatus and medium

This invention discloses a training method, conversion method, apparatus, device, and medium for a language conversion model, belonging to the field of computer technology, and solves the problem of high training difficulty caused by the complexity of the model structure. The method includes: concatenating column name information from a natural language query sequence and a semantic matching association table into a long sequence; training a pre-constructed language conversion model based on the long sequence information to obtain a trained language conversion model. The language conversion model is used to determine the correspondence between the natural language sequence and the structured query language sequence, which includes selection clauses and conditional clauses. During the model training process, a first decoder is used to determine the prediction result of the selection clause based on its features, and a second decoder is used to obtain the prediction result of the conditional clause based on its features. This method can reduce the difficulty of model training.
Owner:CHINA UNITED NETWORK COMM GRP CO LTD

Knowledge graph embedding method based on semantic and relational structure fusion embedding

The present invention utilizes the role of semantic relevance and relational structure relevance in association relationship reasoning, and proposes a knowledge graph embedding method based on the fusion of semantics and relational structure embedding. It is suitable for the embedding and reasoning of all knowledge graphs that contain both entity descriptions and association relationships. Compared with the knowledge graph embedding methods based purely on semantic embedding and purely on relational structure embedding, the present invention has good performance in knowledge graph association prediction tasks, entity prediction tasks and unknown entity sample prediction tasks.
Owner:BEIJING UNIV OF TECH

Method, device, equipment and readable storage medium for generating general data relation structure

The present application provides a method, device, equipment and computer-readable storage medium for generating a general data relationship structure. The method for generating a general data relationship structure includes: automatically extracting entity information from a data table; using a model algorithm to normalize entity information; extracting commonalities and labeling semantic tags for entity information after entity normalization; extracting entity relationships between entities; storing mapping relationships between data tables and entities, mapping relationships between entities and semantic tags, and entity relationships; generating entity relationship graphs; and generating data table relationship graphs. The present application can handle situations where the relationship data of target data is stored in multiple two-dimensional tables, adding more analyzable scenarios to data analysis, and the structure supports drill-down operations. The target data's relationship can be used to find its related data, and the next step of related data query can be performed based on the related data, so that the relationship data is connected to form an effective data network structure.
Owner:BEIJING YUANIAN TECH CO LTD

Data kinship management method, system and storage medium

An embodiment of the present invention discloses a data lineage relationship management method, system and storage medium, which can reduce the management complexity of data lineage relationships and improve query efficiency. The method may include: a data construction parser; a syntax tree construction parser connected to the data construction parser; a graph database module connected to the syntax tree construction parser; wherein the data construction parser is used to parse the data source change data sent from the business system to obtain data lineage relationship structure data; the syntax tree construction parser is used to convert the data lineage relationship structure data into a multi-layer syntax tree instance adapted to the graph database module; the graph database module is used to store and manage graph structure data corresponding to the multi-layer syntax tree instance, the graph structure data is used to represent the graph attribute data of each node in the multi-layer syntax tree instance and the data lineage relationship path between different nodes; the graph attribute data includes the data table and / or data table field corresponding to each node.
Owner:CHINA MOBILE (SUZHOU) SOFTWARE TECH CO LTD +1

A multimodal patent semantic retrieval method and system

This application belongs to the field of information retrieval technology and discloses a multimodal patent semantic retrieval method and system. By establishing relation anchor points, extracting local cross-modal relation representations around the anchor points, and verifying the cross-modal consistency of these relations, reliable cross-modal relations can be identified. Furthermore, technical features / components and successfully aligned cross-modal relations are constructed into a graph structure representation, and a graph matching algorithm is used to calculate similarity. As a result, the overall logic and details of the technical solution can be understood more effectively, and solutions containing only the same technical features but with different relational structures can be distinguished, thereby improving the accuracy and relevance of multimodal patent retrieval.
Owner:SHENZHEN MUNICIPAL CUBE TECHNOLOGY CO LTD

Multi-source heterogeneous data automatic fusion method based on SQL generation

The invention discloses a multi-source heterogeneous data automatic fusion method based on SQL (Structured Query Language) generation, which comprises the following steps of: extracting a relation structure between a question word and a pattern item through a pre-training language model, and constructing an input heterogeneous graph in combination with a predefined relation structure; embedding nodes in the heterogeneous graph by using a pre-training model to obtain a node embedding matrix, and calculating edges in the heterogeneous graph through a syntactic analyzer and judging a relation type to obtain a relation embedding matrix; coding the node embedding matrix and the relation embedding matrix through a relation graph attention network; an LSTM network is adopted as a decoder to decode the output vector matrix of the nodes in the relational graph attention network, and a prediction result is calculated; and extracting and fusing multi-source heterogeneous data according to the generated SQL statement query, and presenting a result. According to the method, the SQL generation accuracy can be improved, and multi-source data fusion is simpler.
Owner:SHIJIAZHUANG TIEDAO UNIV

Type conflict detection method and device, equipment and storage medium

The embodiment of the invention provides a type conflict detection method and device, equipment and a storage medium. The method is applied to the technical field of computers and comprises the steps that a security policy file of a program is configured according to AVC alarm information of the program, and then a policy set with newly-added security policy configuration is obtained; extracting a TE rule set in the policy set after the new security policy configuration; converting a node relationship in the TE rule set into a graph structure to obtain a relationship structure type among nodes in the TE rule set; and judging the relationship between the type corresponding to the newly added security policy and the type corresponding to the original security policy according to the relationship structure type. In this way, a user can judge whether the type corresponding to the newly added security policy conflicts with the type corresponding to the original security policy when adding the security policy, so that the user can conveniently and efficiently configure the security policy file of the program, the configuration cost is reduced, and the security of software development is improved.
Owner:LERONG SMART HOME (BEIJING) TECHNOLOGY DEVELOPMENT CO LTD

Model fine tuning method and system for integrated classifier multiplexing

The invention discloses an integrated classifier multiplexing model fine tuning method and system, belongs to the technical field of artificial intelligence, and aims to solve the problems of feature structure damage and generalization ability reduction caused by hard masks after pruning of an existing model. The method comprises the following steps: constructing a virtual teacher model in a self-adaptive weighting manner based on the accuracy rates of local models of a plurality of clients on a public verification set; taking the pruned global model as a student model, performing multi-level knowledge distillation fine tuning by using a public data set, and covering output logic value alignment, middle layer feature map alignment and inter-class relation structure alignment; and repairing pruning damage through combined optimization of a composite loss function. The system comprises a data and model acquisition module, a virtual teacher model construction module, a composite loss calculation module and a parameter optimization module. According to the method, the performance and generalization ability of the pruning model can be effectively improved, and collaborative optimization of compression and precision is realized.
Owner:UNIV OF ELECTRONICS SCI & TECH OF CHINA