98 results about "Call graph" patented technology

The invention discloses a vulnerability hidden danger intelligent detection method based on a large model, and the method comprises the steps: firstly carrying out the global static analysis of a source code set, constructing a complete call graph and a complete data flow graph of a program, and forming a structured code knowledge graph; and then, aiming at the identified candidate vulnerability slices, based on the maps, carrying out accurate context retrieval and enhancement, converting key information such as a call chain and a data traceability path which are strongly related to the vulnerability slices into natural language description which can be understood by a large language model, and injecting the natural language description into cue words, so that missing global context information is provided for the model. And the defect of complex code analysis capability is overcome. In this way, the problem that an attention mechanism loses efficacy in remote code association is solved, and the accuracy and reliability of vulnerability detection are remarkably improved.

The invention belongs to the technical field of computers, and particularly relates to a software fault repairing method and system fused with intelligent analysis, which comprises the steps of collecting a multi-level running log and performing structured preprocessing, constructing a dynamic calling graph through a time sequence encoder and a graph neural network, inferring a fault root cause in combination with a Bayesian causal inference model, and repairing a fault fault according to the fault root cause. And matching the repair strategy to generate an atomization instruction sequence, and deploying the atomization instruction sequence to a production system after sandbox environment verification. The system comprises a log acquisition module, a feature coding module, a graph construction module, a causal reasoning module, a strategy matching module, an instruction generation module, a sandbox verification module, a deployment feedback module and the like. Through end-to-end intelligent analysis and a closed loop verification mechanism, the fault positioning precision and the repair safety are remarkably improved, system self-evolution is supported, and operation and maintenance are promoted to be transformed from passive response to active autonomy.

The embodiment of the invention provides a multi-round dialogue security defense method and device, electronic equipment, a storage medium and a program product, and relates to the technical field of data security. According to the method, dialogue semantic graphs capable of representing multi-round dialogue semantic evolution, intention circulation and entity association are constructed by obtaining multi-round dialogue content and a tool calling sequence; the method comprises the following steps of: respectively inputting user input information into two types of maps according to a tool calling graph representing a tool calling time sequence relationship and an interaction relationship to extract double risk feature vectors based on multi-round contexts and tool calling logic, and carrying out collaborative risk judgment in combination with a preset judgment rule; according to the invention, rapid integration of multi-round context information and accurate identification of attack risks are realized, so that a protection strategy can adapt to dynamic interaction requirements of an intelligent dialogue system in time, and the technical problem of low defense efficiency in the prior art is effectively solved.

The invention relates to the field of interface component control, in particular to an interface component control method and device, electronic equipment and a storage medium. The method comprises the following steps: acquiring a calling relationship of all interface components, and constructing a component calling graph based on the calling relationship; constructing a main function of the interface component according to the component call graph; when the obtained control command is a starting instruction of a to-be-started interface component, starting of the to-be-started interface component is controlled according to the starting command and a main function of the to-be-started interface component; when the obtained control instruction is a closing instruction of a to-be-closed interface component, closing of the to-be-closed interface component is controlled according to the closing instruction, the component calling graph and a main function of the to-be-closed interface component. According to the method, the multi-level interface component can be accurately controlled, and the display observability and logic control accuracy of the interface component are improved.

The invention provides a code review method and system based on a multi-dimensional knowledge graph and electronic equipment. The method comprises the following steps: constructing a business knowledge graph, a method calling graph and a code review knowledge graph; code data are obtained, a target code in the code data is determined based on the code review request, and the target code comprises code change content; generating a large model cue word in combination with the business knowledge graph, the method calling graph and the code review knowledge graph; and according to the big model cue word, calling an AI big model to perform joint review on the code change content, and generating a review report which at least comprises a repair suggestion. According to the method, multi-dimensional knowledge of business, code calling and code review is integrated, a more comprehensive and closely associated knowledge system is formed, the generated cue word can accurately guide the AI large model to review code change in combination with multi-field knowledge, blindness of the large model during review is avoided, and review efficiency and accuracy are improved.

An impact analysis device acquires a list of method information related to a difference between a new version of a library and an old version of the library, based on a source code of the new version of the library and a source code of the old version of the library; extracts, from the list, a first difference related to a method existing in a first call graph of software that uses the old version of the library; extracts, from the list, a second difference related to a method added to a second call graph of the software that uses the new version of the library; extracts a set of method information of a caller of a transition added to the second call graph in comparison with the first call graph; and outputs a union of the first difference, the second difference, and the set, thereby enhancing support for determining a likelihood that a library update could affect operations of software.

A computer-implemented method for trace-driven call-graph-aware proactive coordinated autoscaling of component microservices in an application includes generating performance-resource elasticity models of endpoints of the component microservices of the application. Workload levels of the endpoint of the component microservices is predicted based on user traffic observed at a front end service. A trace-level performance of the application is predicted for different microservice replica scaling based on the performance-resource elasticity models at end points, the ends points on the trace call graph and the predicted workload levels. A microservice replica scaling is recommended for each of the component microservices to meet predefined trace-level user service level objectives.

The invention discloses an intelligent detection method for network security vulnerabilities, and relates to the technical field of network security vulnerability detection, and the method comprises the following steps: analyzing source codes to construct an abstract syntax tree, a control flow graph and a call graph, identifying sensitive operation points based on a rule base, and screening hotspot functions in combination with code indexes and the call graph; extracting a candidate path from a program entry to a sensitive operation point, encoding to generate a path context snapshot, and scoring through a pre-training model; selecting a high-risk path to execute sparse symbolic execution, simplifying constraints by using variable interval information, and dynamically feeding back and adjusting scores; poC input verification is generated for the path triggering the vulnerability constraint, a report containing the path constraint, the triggering condition and the PoC is output, program analysis and machine learning are combined, the vulnerability detection efficiency and precision are improved, and the method is suitable for code security analysis of scenes such as an embedded system.

The invention discloses a method and a device for generating a software vulnerability analysis report, and relates to the field of information security. The method comprises the following steps: acquiring auditing information of software vulnerabilities; a code feature matching rule for the software vulnerability is generated according to the auditing information through the large language model, and the code feature matching rule is used for matching a code calling mode related to the software vulnerability in the target software project; constructing a calling graph of the target software project; according to the code feature matching rule and the call graph, the correlation degree between each code snippet of the target software project and the software vulnerability is detected, and the correlation degree is used for evaluating the risk level of each code snippet of the target software project; and generating a software vulnerability analysis report of the target software project according to the association degree between each code snippet of the target software project and the software vulnerability. The technical problem that in the prior art, an open source software vulnerability detection tool cannot accurately evaluate the real influence of vulnerabilities in specific project contexts is solved.

The invention discloses a fault propagation path determination method and device, a storage medium and electronic equipment, and relates to the field of Internet of Things. The method comprises the following steps: acquiring a target data set; a service calling graph is constructed according to the target data set, then strong connectivity vectors in the service calling graph are determined, and the service calling graph is used for representing a topological structure and a dependency relationship of service interaction in the distributed system; the strong connectivity vector is used for representing a connectivity component of which the bidirectional calling frequency is greater than a preset frequency and / or the data interaction strength is greater than a preset strength in the service calling graph; performing neighborhood feature aggregation on the service call graph to obtain a time sequence feature matrix; and determining a fault propagation path of the distributed system according to the strong connectivity vector and the time sequence characteristic matrix. According to the method and the device, the technical problem of poor fault propagation path prediction accuracy caused by lack of quantification of dependency relationship and topological association between services in fault propagation prediction in the prior art is solved.

The invention discloses a code abstract generation method and system based on hierarchical context awareness, and belongs to the field of natural language processing and software engineering. The method comprises the following steps: generating hierarchical annotations for a code warehouse based on a function call graph, firstly extracting a call relationship in the graph and eliminating a ring structure, and then classifying leaf nodes and generating annotations; a topological sequence from bottom to top is adopted in the core step, the importance score of a child node is calculated for each father node, child node annotations, class function descriptions and code bodies of the child nodes are fused, and a large language model is input to generate annotations till the whole graph is covered. The method is suitable for an automatic document generation scene of a complex software project, and through a technical path combining function call graph analysis, child node importance measurement and class function description, the accuracy and semantic consistency of code annotation are improved; the method can be widely applied to the fields of software maintenance, code understanding, intelligent development tools, automatic document generation of enterprise-level code warehouses and the like.

The invention discloses a file code similarity analysis method, which is based on a directed graph isomorphism theory and comprises the following steps of: A, extracting characteristics of an open source code file, namely a Call Graph (Call Graph for short), and establishing a file sample library; b, extracting a function call relation graph of the file to be analyzed; c, executing the standardization work of the graph, and preprocessing the graph according to the related definition of the tree structure; d, extracting the maximum common subgraph of the function call relation graph of the file to be analyzed and the sample library file; and E, calculating a relationship between the maximum common subgraph and the function call relationship graph to obtain a similar result, and completing file code similarity analysis. According to the method, the similarity of the file codes is analyzed by depending on the function call relation graph, the internal characteristics and higher-level logic characteristics of the file codes are considered, and higher accuracy and efficiency are achieved. The method is suitable for occasions such as source code plagiarism detection.

The embodiment of the invention provides a test method and device for an operating system kernel based on a Rust language, and the method comprises the steps: obtaining a source code of the operating system kernel based on the Rust language, compiling the source code into an intermediate representation, and generating a function call graph among functions contained in the operating system kernel according to the intermediate representation; determining a target position corresponding to a preset potential anomaly in the intermediate representation and a target variable corresponding to the potential anomaly in the target position through a preset recognition rule; performing backward tracking analysis on the target variable according to the function call graph and the target position, and determining a system call interface associated with the target position in an operating system kernel and a call variable associated with the target variable in the system call interface; and according to the system call interface and the call variable, generating a test case for the operating system kernel.

The invention relates to the field of security vulnerability early warning, and particularly discloses a software security vulnerability early warning method based on deep learning, which comprises the following steps: starting from a predefined sensitive sink, screening out a code path set directly related to a potential risk by constructing a global function call graph and carrying out call chain reverse tracing; and the analysis redundancy is reduced from the source. Furthermore, according to the scheme, simple sequence modeling is not carried out on the call chain, function nodes in the call chain are converted into deep semantic vectors, and the semantic influence of key nodes is amplified and the interference of irrelevant information is inhibited by dynamically evaluating the time sequence importance of each function node in the call chain and through adaptive weighting. And finally, deep learning is carried out on the key path subjected to information discrimination, and vulnerability classification prediction is carried out, so that the problem of information overload in a long call chain is effectively overcome, and the accuracy of vulnerability early warning and the capture capability of complex semantic association are improved.

The invention relates to the technical field of recognition for open set shelled software, in particular to a recognition method, system, equipment and medium for open set shelled software, which comprises the following steps of: extracting function call graph structure information from a binary executable file to be recognized through a disassembling tool and a user-defined plug-in thereof; generating a structured graph representation file; on the basis of the graph representation file, graph statistical features and file section structure features are extracted, and a spliced feature vector is formed; and based on a multi-model single-class identification strategy constructed for open set identification, carrying out classification identification on the binary executable file corresponding to the current feature vector. The method has the beneficial effects that the identification efficiency of the shelled software is integrally improved, the detection capability of the unknown type of shelled software is particularly enhanced, and powerful support is provided for dynamic malicious software analysis.

The application discloses a kind of based on big language model's evolvable software vulnerability detection method, comprising: by regular pattern matching identification Source sentence, based on call graph traversal and data dependence analysis execution function level inter-process slice, build cross-function code context, input the big language model of parameter efficient fine-tuning, output the vulnerability propagation path from Source to Sink;When new vulnerability type needs to be extended, the parameter variation characteristics of old data are extracted by multi-step fine-tuning, and the representative core set is selected by random projection dimension reduction and hybrid distance hierarchical clustering, and the training is played back by mixing new data to alleviate catastrophic forgetting;In the actual use process of tool, the false alarm and the false alarm confirmed by user are collected as feedback signal, the core set is clustered and layered filtered and refined based on perplexity and error prediction analysis, the harmful old knowledge that leads to false alarm and false alarm is removed, and verified new mode is supplemented at the same time, to realize the closed-loop evolution of self-improvement.

The application relates to the technical field of software test automation, and provides a requirement-driven test case automatic generation method and system based on a semantic test graph, which comprises the following steps: performing multidimensional static analysis on source code, respectively constructing an abstract syntax tree, a control flow graph and a function call graph, and extracting function call relations and data flow dependency relations; constructing a semantic test graph by taking functions as nodes, taking function call relations as main edges and taking data flow dependency relations as auxiliary edges, and aggregating function nodes in the semantic test graph into several function modules; obtaining several test requirement nodes, calculating semantic similarity between the test requirement nodes and the function modules, establishing a mapping relationship from the test requirement nodes to the function modules, and forming a requirement-program structure association graph; and based on the requirement-program structure association graph, generating a test intention for each test requirement node, and converting the test intention into an executable test case. The highly corresponding relationship between the generated test case and the business requirement is ensured.

This invention provides a permission-based Android malware detection method and system, relating to the field of Android malware detection. The method includes: extracting multimodal features of each function from the APK installation package file of the target Android software and constructing a call graph between functions. The multimodal features include permissions, and the call graph uses nodes to represent functions and edges to represent the call relationships between functions. The multimodal features and call graph of all functions are input into a trained detection model for malicious and benign classification, yielding Android malware detection results. This invention exhibits higher detection accuracy and robustness in static analysis-based malware detection tasks, effectively improving security protection capabilities in practical applications.

The application relates to a code vulnerability detection method and relates to the technical field of computers.The method comprises the following steps: obtaining to-be-detected code; performing preprocessing on the to-be-detected code to obtain a normalized source code representation; constructing an abstract syntax tree based on the source code representation; constructing an intermediate representation based on the abstract syntax tree; wherein the intermediate representation comprises a core expression set, a target control flow graph and a target function call graph; constructing a target value dependency graph based on the target control flow graph and the target function call graph; performing constraint extraction based on the target value dependency graph and the target control flow graph to obtain a target constraint expression; and inputting the target constraint expression into a constraint solver to obtain a vulnerability detection result.The code vulnerability detection method can realize high-precision, low-false-alarm and strong-context-aware code vulnerability detection, and is more suitable for scenes with strict security requirements.

The application discloses a second-order security vulnerability detection method for Java database interaction application, and comprises the following steps: S1, setting a source point, a sink point and a pollution propagation rule of a pollution analysis according to a category of the second-order security vulnerability; S2, processing a target program to be detected to obtain a static analysis IR; S3, combining a configuration file related to database interaction in a target program code, performing a pointing analysis on the IR generated in the step S2 to generate a pointing flow graph (PFG) and a method call graph (CG) connected with the Java application and the database; and S4, performing a pollution analysis on the IR generated in the step S2 based on the pollution analysis rule set in the step S1 and the pointing flow graph and the method call graph obtained in the step S3. Compared with a traditional vulnerability detection method for the Java application, the application collects a pollution flow between the Java application and the database by constructing a static analysis channel between the Java application and the database, so that the second-order security vulnerability detection is realized.

The invention discloses an LLM-based fuzzy test initial seed generation method and system, a computer and a storage medium. Lightweight static analysis is performed on target software source codes, an expansion function call graph is constructed, two key features of hard coding constraint and security sensitive operation are extracted, and explicit tags are injected in situ into the source codes; a semantic aggregation strategy based on a recursive sub-function abstract is adopted, the sub-function source code injected with the label is converted into a semantic abstract from bottom to top through a large language model, and a global semantic abstract retaining key constraint information is generated through hierarchical cue word construction; constructing a seed generation task containing a thinking chain guide instruction, forcing the model to perform reasoning according to logic steps of constraint solution, abnormal construction and enhanced generation, and generating a targeted initial input sample (namely, an initial seed) for fuzzy testing; and performing static validity check on the grammar by utilizing a general format analyzer, and performing fault-tolerant processing by combining a multi-stage backspacing mechanism. Through the recursive abstract aggregation and thinking chain reasoning technology, deep semantic understanding of complex code logic and automatic generation of high-coverage-rate test cases are achieved.

The application discloses a static program open source component analysis method, relates to the technical field of software analysis, and comprises the following steps: inputting project source code, parsing, obtaining a semantic call graph, and performing hierarchical management on the semantic call graph; performing topological sorting on the hierarchical semantic call graph, obtaining a component inspection sequence; comparing the semantic call graphs corresponding to the new and old versions of a to-be-upgraded component, performing a software detection operation on the to-be-upgraded component, and outputting a conflict report. The method constructs a hierarchical semantic call graph, which contains rich context information, so that analysis can be deepened to the level of whether the behavior and semantics of a method change; through topological sorting on component dependency relations, the application proposes a component inspection sequence, which solves the circular dilemma that repairing one conflict causes another conflict in reality, guarantees that the components depended on by any component to be inspected have been analyzed and confirmed before the component is inspected, and ensures the orderliness and certainty of the analysis process.

The invention discloses a Rust ecological CVE influence tracking method and system based on reverse dependency propagation and function call graph analysis. The method comprises the following steps: connecting a crate.io structured database by taking affected crates and version constraints as starting points, selecting a boundary version meeting a semantic version condition as an analysis source, downloading according to levels by adopting a breadth-first reverse dependency traversal strategy, and unpacking a dependency version in an isolated working area, a parent dependency is automatically injected in a local copy form, and an accurate version is locked through a dependency coverage mechanism of a configuration file, so that reproducibility and offline robust analysis of a revoked version are ensured. The method also supports independent operation of CSV batch task processing and statistics, and has automatic log separation and process-level result traceability. According to the method, the supply chain vulnerability propagation analysis complexity is reduced, and the reproducibility, observability and performance of analysis are improved.

The embodiment of the invention provides an interface detection method. According to the scheme, the method calling graph is constructed according to calling relations among service nodes, and the service nodes comprise method nodes and interface nodes; determining artificial intelligence service method nodes in the method nodes based on preset auxiliary identification code information; and determining a target interface node in the method calling graph by using the calling relationship between the artificial intelligence service method node and each service node, and determining a service interface corresponding to the target interface node as an artificial intelligence service interface.

This invention discloses a malware detection method based on adaptive node-edge joint masking. First, a function call graph of an Android application is constructed, and the structural and semantic features of nodes and edges are extracted. Then, a masking strategy is adaptively determined based on the relative importance of nodes and edges in the graph structure, and joint masking is applied to nodes and edges. On this basis, a graph neural network model is trained, enabling the model to learn stable and discriminative representations even when some graph structures or features are damaged. Finally, the trained model is used to detect malware in Android applications. This invention demonstrates effectiveness in improving the robustness and engineering practicality of Android malware detection models.

The invention belongs to the technical field of information security, and particularly relates to an Android malicious software detection method based on enhanced operation code semanteme, which comprises the following steps: performing static analysis on an Android application sample to obtain a function call graph and an operation code sequence; performing sensitive API screening on the function call graph to obtain a simplified API call graph; assigning an operation code in the operation code sequence of each node in the simplified API call graph according to the relation between the weight of the node to which the operation code belongs and the weight of the whole API call graph; inputting the node vector representation into a Bi-LSTM model to obtain a node time sequence vector representation; clustering based on the node time sequence vector representation to obtain enhanced feature representation; taking the simplified API calling graph represented by the embedded enhanced features as input, taking a malicious software detection result as output, and training a graph convolutional neural network; inputting the enhanced feature representation of the to-be-detected Android application and the simplified API call graph into the detection model to obtain a detection result; according to the method, the problem that the detection model performs deep semantic extraction on the graph is solved, and the detection accuracy is improved.

The invention discloses an operable code odor fusion detection method and system based on a large language model and a multi-index call graph. The method comprises the following steps: firstly, collecting potential code peculiar smell in an open source project, judging the operability according to the evolution history of the code peculiar smell, constructing a code calling graph, and constructing a code peculiar smell data set; secondly, finely adjusting the large language model, extracting a middle hidden layer state and generating a code semantic vector; embedding the static code measurement indexes into nodes of the code calling graph to form a multi-index calling graph, and representing code structure vectors by using a graph neural network with an attention mechanism; and finally, carrying out progressive feature fusion on the code semantic vector and the structure vector, and detecting whether the code contains an operable code peculiar smell or not through a multi-layer classification network. According to the method, the code semantic features are obtained by using the fine-tuned large language model, the code structural features are obtained by relating to the multi-index calling graph, and the operable code odor detection capability is further improved by representing the code features.

The invention discloses a decision logic analysis method and related equipment, and the method comprises the steps: collecting a real-time rule trigger event flow and a decision logic source code of a server in parallel when a decision engine of the server runs; constructing a dynamic call graph based on the real-time rule trigger event flow, wherein the dynamic call graph comprises rule nodes and dynamic attributes triggered by rules; constructing a static dependency graph based on the decision logic source code, wherein the static dependency graph comprises a dependency relationship between variable nodes and a code level; analyzing an association relationship between the dynamic call graph and the static dependency graph, and generating a dynamic and static fusion strategy graph; and rendering a thermodynamic diagram based on the dynamic and static fusion strategy map, so as to display the thermodynamic diagram on a visual interface.

This invention relates to the field of artificial intelligence technology and discloses a data processing method, apparatus, device, and storage medium for business processes. The method includes: semantically registering each basic tool to obtain a tool capability catalog; parsing the target business process text to obtain a skill script, where the target business process text contains the business process; compiling the skill script based on the tool capability catalog to obtain a tool call graph; wherein, basic tools serve as nodes in the tool call graph, and the data flow relationships between basic tools serve as edges in the tool call graph; integrating information based on the security attribute information of the nodes to obtain a skill-level capability contract text used to constrain the business process; and executing the business process based on the tool call graph and the skill-level capability contract text. This method can be applied to process processing scenarios in fintech and healthcare, improving the execution efficiency of business processes.

The application discloses a vulnerability risk intelligent detection method based on a large model, which firstly performs global static analysis on a source code set, constructs a complete calling graph and a data flow graph of a program, and forms a structured code knowledge graph; then, for the identified candidate vulnerability slices, accurate context retrieval and enhancement are performed based on the graph, key information such as a strong correlation chain and a data trace path of the vulnerability slice are converted into natural language descriptions understandable by a large language model, and prompt words are injected, so that the model is provided with missing global context information, and the deficiency of the model in complex code analysis capability is made up. In this way, the problem that the attention mechanism is invalid in long-distance code association is solved, and the accuracy and reliability of vulnerability detection are significantly improved.