67 results about "Rule system" patented technology

The invention provides a work fund compliance auditing method and system based on a big language model, and relates to the technical field of compliance auditing, the method comprises the following steps: extracting an unstructured text, and using the big language model to carry out data cleaning, privacy desensitization and standardization processing; based on a large language model and a retrieval enhancement generation technology, constructing a work fund compliance knowledge graph, performing deep auditing analysis on standardized data, and identifying income and expenditure anomalies, asset vulnerabilities and compliance risks; building a compliance rule base by using a large language model, and performing automatic compliance judgment; generating a differentiated improvement suggestion, a rectification path and an audit report first draft, and continuously optimizing a model and a rule; the system comprises a multi-source data acquisition and preprocessing module, a deep audit analysis module, a compliance analysis module, a decision support and report generation module, and a learning and optimization module. According to the invention, through automation and intelligentization of the work fund auditing process, the auditing efficiency, the auditing accuracy and the self-adaptive capability are remarkably improved.

The invention provides an intelligent time sequence security analysis method and device based on a large language model, and relates to the technical field of information security, and the method comprises the steps: collecting an original log from a multi-source heterogeneous system to obtain a standardized log event; inputting the standardized log event into a preset large language model to generate a historical semantic log event; constructing a user digital portrait based on the historical semantic log event, and establishing a dynamic behavior baseline model associated with the user digital portrait by using a long-short term memory neural network model; obtaining a semantic log event stream output by the large language model in real time, and querying a dynamic behavior baseline model of a user corresponding to the semantic log event stream; and evaluating whether the current semantic log event is an abnormal event: if the current semantic log event is an abnormal event, calling a large language model to generate a semantic alarm, and injecting the semantic alarm into a rule engine. The technical problems that a traditional static rule system is poor in adaptability, high in false alarm rate and high in maintenance cost are effectively solved.

A water level data anomaly detection method based on rule and neural network fusion comprises the following steps that S10, upstream and downstream water level stations associated with a target water level station are determined, and rainfall stations influencing the water level of the water level stations are analyzed; s20, converting the collected and processed data into a time sequence format suitable for the input of the LSTM model; s30, constructing an anomaly judgment rule system of the target monitoring station, converting a rule into a differentiable mathematical constraint term, and embedding a training target loss function of the LSTM model; s40, a rule-constrained LSTM model is constructed and trained, whether the model is converged or not is judged based on the change trend of a target loss function, and model parameters and rule weight coefficients are corrected; and S50, using the trained rule to constrain the LSTM model, and performing water level abnormal data detection and result output. According to the method, the rule parameters are fused into the LSTM, two normal forms of data driving and knowledge driving are combined, the abnormal water level data are accurately recognized, and meanwhile, the interpretability and engineering applicability of an abnormal detection result are improved.

The invention discloses a sheet metal part manufacturability reasoning method based on space-semantic map alignment, and relates to the field of manufacturing-oriented design evaluation and industrial knowledge reasoning, and the method comprises the following steps: carrying out geometric analysis on a CAD geometric model of a sheet metal part to be evaluated; abstracting the geometric features and the topological / metric spatial relationship thereof into a computable spatial semantic graph; performing semantic analysis on the process specification described by a natural language, converting the process rule into formalized logic check expression by using a large language model through context learning, and generating an executable domain-specific language check script; executing the script on a spatial semantic graph, realizing deterministic reasoning through graph matching and attribute verification, and completing accurate mapping and violation detection of text rules and geometric features; and outputting an interpretable diagnosis result containing violation feature positioning, triggering rules and numerical evidence. In order to solve the problems that a process rule'natural language-geometric model 'has a semantic gap, a traditional rule system is poor in adaptability, and an end-to-end learning method is high in data dependence and cannot be explained, a new rule can be quickly adapted under the condition that a large amount of data does not need to be labeled and a model does not need to be retrained; the method can accurately identify the violation of the micro-size and spatial relationship, and has reasoning preciseness, interpretability and engineering availability.

The invention discloses a personal data value management system and method. The system comprises a main body collaboration layer which is used for defining roles and responsibilities of multiple main bodies and constructing a multi-main-body collaboration framework; the process management and control layer is used for constructing a full-process closed-loop design and establishing linkage rules of all links; the rule guarantee layer is used for constructing a rule system and dynamically adjusting the rule system according to scenes, policies and markets; the value distribution layer is used for carrying out quantitative processing on the individual contribution degree and the enterprise contribution degree and carrying out income distribution according to the individual contribution degree and the enterprise contribution degree; and the personal participation layer is used for constructing a personal participation mechanism so as to obtain personal data dynamics and rights and interests in real time. By adopting the system provided by the embodiment of the invention, a five-layer cooperative personal data valued management system is constructed, each layer not only defines the core function, but also realizes cooperation through data synchronization, rule linkage and a subject counterbalance mechanism, and a management closed loop covering the full life cycle of personal data is formed.

The invention discloses an intelligent form processing method and system based on rule tree optimization and large model screening, and relates to the technical field of artificial intelligence and data automation. The problems that the processing efficiency is low due to static solidification of a rule system, redundancy of execution paths and dispersion of homogeneous rules in a traditional form processing method, and the screening accuracy is low and the labor cost is high due to the lack of effective judgment capability for complex, fuzzy or implicit condition data of semantic understanding are solved. By constructing and optimizing the rule tree structure and combining with the semantic comprehension ability of the large language model, automatic grading processing and intelligent screening of the form are realized. Firstly, deterministic rule matching and filtering are completed by a rule tree, then semantic judgment is carried out in a logic context by utilizing a large model for uncertainty data, and finally, a screening result is fused and output. According to the method, the efficiency, the accuracy and the automation level of form processing in a complex business scene are remarkably improved.

This invention belongs to the field of artificial intelligence technology, specifically disclosing an intelligent rule conflict detection method, system, and program product. By transforming the traditionally tedious and time-consuming manual rule review into intelligent rule conflict detection, it greatly simplifies the verification process for business personnel in rule design and significantly improves the efficiency of rule review. It can comprehensively and deeply analyze the rule system, automatically identifying logical errors, overlaps, and contradictions that are difficult for the human eye to detect, effectively reducing the risk of rule defects and potential financial losses caused by human negligence. It deeply embeds artificial intelligence technology into the business operation process of rule configuration, ensuring the logical validity of rules once they are created, and promoting the upgrade of enterprise rule management from reliance on manual experience to intelligent automatic detection. Furthermore, it can intuitively display potential problems and conflicts between rules, making the entire rule verification process clearer and easier for users to understand and make decisions.

The invention discloses a production scheme selection method, device and system, a storage medium and a computer program product. The method comprises the steps of obtaining multi-dimensional production data in a production process; processing the multi-dimensional production data to remove invalid data and unify data formats; target machine type information is received, and at least one corresponding initial production scheme is called from a preset production scheme library according to the target machine type information; based on a preset rule system which can be dynamically adjusted independent of a business process and contains inter-rule association logic and scenarized rule grouping, performing matching screening on the multi-dimensional production data and the initial production scheme to obtain a candidate production scheme; performing quantitative evaluation on the candidate production schemes by adopting a preset multi-dimensional evaluation index system, and determining an optimal production scheme; and verifying the optimal production scheme, and outputting the scheme after the verification is passed. According to the scheme, the flexibility and maintainability of the scheme selection technology are improved, and the scheme selection accuracy is improved.

The invention relates to the technical field of artificial intelligence, in particular to a papermaking employee virtual practical training method and system combined with a large model. According to the papermaking employee virtual practical training method and system combined with the large model, immersive interaction with a high-fidelity virtual papermaking production line is achieved through data collection, virtual simulation and large model interaction and reasoning. According to the method, a large-scale language model is introduced, so that the method has strong natural language understanding, deep reasoning and multi-modal information fusion capabilities, complex and changeable virtual environment states can be analyzed in real time, and intelligent diagnosis and guidance are performed in combination with natural language questions of employees; and even non-preset or multi-factor coupled fault scenes which cannot be processed by a traditional rule system can be processed, man-machine dialogue interaction similar to experienced experts is realized, and the intelligent level of the system on solving complex problems is remarkably improved.

The application discloses a kind of comprehensive data analysis system and method applied to intelligent audit, it is related to big data analysis technical field, the present application is by obtaining audit object whole-link business data, constructs full audit data set by standardization processing;Extract static and dynamic risk characteristics, build the comprehensive rule system containing static and dynamic required verification data type library;Analysis single risk and associated business link causality, generate audit associated link mapping table and ordered risk set;Calculate corresponding risk determination threshold;Real-time monitoring audit business based on threshold, data characteristics are intercepted by sliding window, after audit cycle ends, iteration updates threshold.The system includes data full collection, audit rule construction, associated link analysis, determination threshold calculation, audit threshold iteration module and subordinate unit.The present application realizes full data coverage and real-time monitoring, accurately traces risk transmission, improves the confidence and timeliness of audit compliance judgment.

Disclosed in the present description are a multivariate decision processing method and apparatus, a storage medium, and an electronic device. The method comprises: acquiring an initial multivariate decision rule system configured by a user end on the basis of standard DSL grammar rule information, wherein the standard DSL grammar rule information is DSL grammar rule information preset for a multivariate transaction decision development scenario; determining all multivariate decision rule combinations corresponding to the initial multivariate decision rule system, and determining reference transaction workflows corresponding to the multivariate decision rule combinations; constructing a multivariate combination instance library for the initial multivariate decision rule system, and on the basis of the multivariate combination instance library, performing engineering language conflict detection processing on the initial multivariate decision rule system, so as to obtain a target multivariate decision rule system after conflict detection; and using the target multivariate decision rule system to perform transaction workflow decision processing on transaction data.

Embodiments of the present application provide a user property safety risk early warning method, device, medium and product, relating to the technical field of communication. The method comprises: based on the user's operator network behavior data, obtaining the user's communication behavior data, location data and application usage data; generating target features according to the communication behavior data, the location data and the application usage data, and constructing an abnormal feature matrix according to the target features; according to the abnormal feature matrix, using a preset statistical generation model, a deep learning discriminant model and a knowledge-driven rule system, respectively generating a first generation result, a second generation result and a third generation result containing the user's risk state, and determining the user's predicted state result based on the three results, and performing corresponding property safety warning behavior based on the predicted state result. The scheme of the present application solves the problem of limitations in the security attack risk of the old people's telecommunication mode.

Systems and methods include causing presentation of a first cluster in association with an event of the first cluster, the first cluster from a first set of clusters of events. Each event includes a time stamp and event data. Based on the presentation of the first cluster, an extraction rule corresponding to the event of the first cluster is received from a user. Similarities in the event data between the events are determined based on the received extraction rule. The events are grouped into a second set of clusters based on the determined similarities. Presentation is caused of a second cluster in association with an event of the second cluster, where the second cluster is from the second set of clusters.

A system can establish a database that includes a plurality of atomic units from documents relating to one or more clinical actions. The system can receive a request for authorization of a clinical action and determine, using a rules engine, one or more rules for a type of the clinical action. The system can generate a query to the relational database to retrieve, from the relational database, a group of atomic units dynamically identified as corresponding to the type of chunk and based on the one or more filters. The system can determine, by the rules engine using the one or more rules, that the group of atomic units resulting from the query satisfy the one or more rules. The system can authorize the clinical action responsive to the determination.

The embodiment of the invention provides an intelligent ship and cargo matching method, device and equipment. The method comprises the following steps: in response to received rule constraint data predefined by a user, constructing and initializing a rule base about ship and cargo matching based on the rule constraint data; wherein the rule base comprises a multi-level rule system which is divided into a hard constraint and a soft constraint; in response to the received ship transport capacity data and cargo demand data, performing preliminary matching screening on the ship and cargo combination based on a rule base, and generating a candidate matching set; based on the candidate matching set, a Pareto optimal solution set for ship and cargo matching is solved and generated through a multi-objective optimization algorithm; in response to the received real-time shipping data, updating a matching scheme in the Pareto optimal solution set through a dynamic feedback mechanism, and generating an updated matching scheme; and determining a final ship and cargo matching result based on the updated matching scheme. By using the method provided by the embodiment of the invention, the decision-making efficiency of shipping logistics can be comprehensively improved.

The invention discloses a data quality auditing and scoring model, which relates to the technical field of data processing and comprises the following steps of demand acquisition and scheme formation, rule system construction, data auditing execution, comprehensive quality scoring and problem closed-loop processing. According to the method, management and control objects, core data items, six data quality dimension standard-reaching requirements and differentiated management and control thresholds are defined in a demand acquisition stage, a data quality scheme is formed, a quality rule system covering beforehand, during and after beforehand is constructed, a verification logic module is formed according to a six-dimension disassembling rule, and a verification result is obtained. A data quality auditing model is built to verify target data, deep linkage of demand-rule-auditing is achieved in the whole process, a rule system and a management and control target are precisely matched, the data quality auditing model executes verification according to a standardized dimension, the dimension problems of data integrity, uniqueness and the like can be precisely positioned, and the verification efficiency is improved. And clear data quality control direction and efficient execution are ensured.

The invention relates to the technical field of interpretable reasoning, in particular to an enterprise AI ethical compliance management platform combined with a knowledge graph, and the platform comprises the steps: analyzing an enterprise ethical file to generate a clause node set, comparing nodes, reconstructing responsibility conflicts to establish a responsibility level graph, calling the graph to carry out semantic dependency mapping on enterprise behaviors and clauses, constructing a behavior ethical mapping set, evaluating the transparency of a mapping path, adjusting a display scheme, generating a reasoning transparent path set, evaluating the compliance according to the path, carrying out the classification identification, and outputting an enterprise ethical compliance management result. According to the method, ordered arrangement of conflict relationships and differentiated definition of rule boundaries are realized through precise disassembly of constraint terms and layered construction of responsibility relationships, structured mapping with hierarchy and dependency features is formed through semantic path matching, the verifiability of logic trajectories is enhanced by using a transparency grading mechanism, and finally, differentiated risk presentation is performed. And the rule system is coherent, explainable and controllable.

The invention discloses a business document cost allocation method and device, equipment and a medium, and the method comprises the steps: obtaining the document type of a business document; based on a plurality of preset extraction fields corresponding to the receipt type, extracting cost information from the business receipt; based on the cost information, determining an apportionment rule of the business document; and based on the apportionment rule and the cost information, generating apportionment result data of the business document. By integrating allocation scenes of different business receipts and constructing a universal and configurable allocation rule system, various business receipts can directly quote a unified rule to realize cost allocation, and the problems that a traditional manual processing mode is low in efficiency and easy to make mistakes are effectively solved; while the allocation accuracy and the execution efficiency are remarkably improved, the workload of repeated development and customization is greatly reduced, and the manpower and cost investment is reduced; and by supporting flexible rule configuration of multiple service scenes, the application flexibility and universality of pre-allocation of the bill cost are further enhanced.

The invention belongs to a text data processing system or method in the field of administrative penalty. The invention discloses an administrative auxiliary quantity penalty method based on a large model, and the method comprises the steps: 1, inputting core data, and providing an environment field case original text, law articles, a case vector database and related basic data; 2, text preprocessing is conducted, non-target domain texts are intercepted through a domain verification-multistage cleaning-structured regularization three-order framework, data noise is eliminated, and structured analysis is completed; 3, quantity penalty core elements are extracted, model context limitation is broken through by adopting a fragmentation-extraction-aggregation framework, and accuracy of quantity penalty elements is ensured by combining multi-round iterative extraction and verification optimization; step 4, quantity penalty amount calculation: depending on an expert rule system and a layered random forest regression model, realizing scene-based quantity penalty amount calculation; and step 5, outputting a result, generating an administrative penalty decision draft, a quantity penalty basis and an amount detail, and attaching a confidence label to ensure that the result is traceable.

According to the semantic rule association system and method based on the accounting knowledge graph, an accounting element three-dimensional semantic modeling system and a rule association degree quantitative evaluation model are introduced, and the problems that accounting elements are isolated and stored, change is complex, and evaluation is difficult in a traditional method are solved; and the universality and high self-evolution of the rule system are realized. The method is suitable for a financial sharing center, a transnational group tax management system, an intelligent auditing platform and other scenes needing to process complex accounting rule association, and is particularly beneficial to prevention of finance and tax compliance risks.

This invention relates to the fields of artificial intelligence and knowledge graph technology, specifically to a knowledge graph inductive processing method based on message block constraints. The method includes: merging messages with the same source node, destination node, and topic into message blocks; constructing a three-layer rule system of structural constraints, semantic constraints, and content consistency constraints to perform multi-dimensional verification of the message blocks, eliminating message blocks that do not conform to the constraint rules; performing hierarchical inductive reasoning on the verified message blocks, generating knowledge graph triples for known entity-relationship, unseen entity, and unseen relation respectively; validating the generated triples, integrating the verified triples into the knowledge graph, and recording the source association information with the source message block. This invention achieves accurate transformation from scattered messages to structured knowledge, effectively improving the accuracy and dynamic adaptability of knowledge graph construction, and can be widely applied to various knowledge graph construction scenarios based on message data.

The application discloses a rule-based lightweight metadata acquisition and analysis method, which comprises the following steps: step one, based on metadata acquisition and analysis requirements, unified modeling and initialization of acquisition rules, acquisition granularity control rules and analysis rules are carried out to form a rule system; step two, unified constraint of metadata acquisition and analysis behavior is carried out through construction of a rule system covering acquisition range, acquisition granularity and feedback adjustment; step three, through acquisition granularity decision and minimum necessary metadata set generation based on historical structure change rate, dynamic adjustment of acquisition scale is carried out according to the structure change degree; step four, through hierarchical weighted structure fingerprint generation and fingerprint difference comparison, metadata structure change is identified; and step five, through the incremental acquisition and rule feedback mechanism, the rule-based lightweight metadata acquisition and analysis method with lightweight, self-adaption and high pertinence operation in the metadata acquisition and analysis process is realized.

This invention relates to the field of rule engine technology, specifically to a method for generating and matching dynamic rule scripts that supports nested contexts and real-time visual editing. The method includes the following steps: extracting and constructing a placeholder script from the nested context skeleton; generating corresponding local semantic patches only for the semantic perturbation range; comparing the placeholder scripts and candidate dynamic rule scripts in terms of hit path, scope traversal order, and reference landing point to determine whether the local semantic patch causes a diffusion offset beyond the semantic perturbation range; if no diffusion offset is caused, finalizing the candidate dynamic rule script as the official dynamic rule script and using it for rule matching; if a diffusion offset is caused, outputting the conflict location result and writing it back to the real-time visual editing interface. This invention avoids the problem of traditional rule systems needing to regenerate the entire set of rule scripts when making local modifications, thus strictly limiting the rule update scope to the affected area.

The invention discloses a low-altitude tourism user portraying method based on journey mapping and multi-modal analysis, relates to the technical field of low-altitude tourism, and solves the technical problem that dynamic cognition and potential intentions of consumers cannot be captured when a traditional rule system or a shallow analysis method faces a low-altitude tourism scene. The method comprises the following steps: S1, identifying a journey stage and marking contact points; s2, performing multi-modal data acquisition and preprocessing according to the divided key stages and key contact points; s3, preliminarily drawing a user portrait according to the preprocessed data in each stage, and verifying the user portrait; s4, outputting a peripheral tag cloud based on the verified multi-layer radar map; s5, determining a brand marketing activity promotion strategy based on the user portrait output in S4; through multi-modal data fusion and time sequence analysis, the method exceeds traditional static portraits, dynamic emotions, cognitive changes and behavior intentions of the user in all stages in a journey are accurately captured, and the microscopic venation of user experience is really understood.

The invention relates to a computational solar photovoltaic building body generation method and system, and belongs to the technical field of building design. The method comprises the following steps: constructing a three-dimensional shape grammar rule system, taking a cube as a basic unit, and formulating various rules based on addition logic; programmed generation and control of the form are realized on a parameterization platform; setting a BIPV performance optimization index, and constructing and generating target guidance in combination with site boundary, density, height and shape constraint conditions; building performance simulation and a multi-objective optimization algorithm are integrated, and a Pareto optimal solution set is obtained; and finally, screening out a building body scheme with excellent performance and form through a multi-dimensional comprehensive evaluation system. The method realizes deep fusion of three-dimensional body generation and BIPV performance optimization, has the advantages of strong expansibility, high coupling, good diversity, flexible control, strong practicability and the like, and is suitable for a building body design scene with BIPV optimization as a guide.

The invention discloses a steel production IF-THEN rule system networking method and device. The method comprises the steps that an IF-THEN rule in the steel production field is disassembled into explicit rule features and original data features; constructing a fusion network comprising an explicit rule path and an implicit data path to respectively maintain rule logic and learning data implicit rules; training the network by adopting a loss function containing a rule constraint term, so that the network fits an original conclusion on a rule triggering sample, and fitting an actual control result on all samples; performing real-time reasoning and decision interpretation by using the trained network; and autonomous optimization is carried out through an incremental learning mechanism after network deployment. According to the method, a traditional rule system is converted into a networked model which can be trained, explained and autonomously evolved, so that the problems of rule expansion and conflict, poor generalization ability, low iteration efficiency and the like are effectively solved, and the intelligent level, the adaptive ability and the decision reliability of steel production control are remarkably improved.

The invention relates to a figure feature recognition and behavior analysis system based on a visual multi-task hybrid model, and belongs to the technical field of computers. According to the system, a local feature extraction unit specifically extracts abnormal behavior related local features, constructs an optimized convolutional neural network model to generate a structured feature vector, and can receive ambiguous feedback to secondarily extract supplementary features; the global behavior reasoning unit distributes computing power through a weight-oriented space-time attention mechanism on the basis of the feature data, and completes multi-type behavior reasoning in combination with scene attribute constraints; the dynamic rule definition unit supports conversion of natural language rules into structured reasoning logic and can dynamically optimize a rule system; and the abnormity early warning closed-loop unit divides early warning levels, triggers response, records abnormal characteristic evolution and generates a backtracking video with a mark. The character behavior recognition accuracy is improved, the rule customization threshold is reduced, a complete early warning tracing closed loop is formed, and the method is adaptive to the multi-scene intelligent management requirement.

Systems, methods, and computer program products are disclosed for scalable event stream processing of trigger rules. Event data associated with a set of object is partitioned into a set of logical partitions, including a first logical partition comprising first event data associated with a first object. The first logical partition is provided to a first event processor to enable the first event processor to update a state associated with the first object based on first event data in the first logical partition. Satisfaction of a first trigger condition specified by a first trigger rule is determined based on the state associated with the first object. A first action is performed in response to the satisfaction of the first trigger condition.

A system can establish a database that includes a plurality of atomic units from documents relating to one or more clinical actions. The system can receive a request for authorization of a clinical action and determine, using a rules engine, one or more rules for a type of the clinical action. The system can generate a query to the relational database to retrieve, from the relational database, a group of atomic units dynamically identified as corresponding to the type of chunk and based on the one or more filters. The system can determine, by the rules engine using the one or more rules, that the group of atomic units resulting from the query satisfy the one or more rules. The system can authorize the clinical action responsive to the determination.

Systems and methods for extracting rule lists from tree ensembles are provided. A system extracts first stage candidate rules from individual trees. The system identifies the first stage candidate rules that satisfy a precision threshold and places those rules in a solution set. Subsequently, a determination is made whether a further stage is needed based on whether a predetermined number of positive data samples of the data set are covered by the solution set. In the further stage, the system generates next stage candidate rules from previous stage candidate rules that have not been pruned and identifies the next stage candidate rules that satisfy the precision threshold, placing those rules in the solution set. A simplified rule list is generated by identifying a minimum subset of rules in the solution set that covers the positive data samples within the precision threshold.