Search

CN-122019097-A - Rule engine data processing method and system based on artificial intelligence and electronic equipment

CN122019097ACN 122019097 ACN122019097 ACN 122019097ACN-122019097-A

Abstract

The invention provides a rule engine data processing method, a system and electronic equipment based on artificial intelligence, relates to the technical field of data processing, and is applied to an enterprise-level automatic data circulation system, wherein the system comprises a multi-level priority association rule and is used for processing multi-dimensional business data of a cross system; the method comprises the steps of inputting unsuccessfully matched data into a pre-trained artificial intelligent model when a high priority rule cannot be matched with part of input data, predicting the associated success rate of the unsuccessfully matched data in a low priority rule, generating a pre-locking instruction if the predicted success rate is lower than a set threshold value, removing the data in advance before executing a subsequent rule to avoid invalid processing, and introducing the artificial intelligent model to conduct pre-prediction and data screening to effectively reduce redundant calculation and improve the matching accuracy and the system resource utilization rate, so that the technical problems of low processing efficiency, poor accuracy and low resource utilization rate in the prior art are solved.

Inventors

  • Ma Chunchuo
  • YU DEMING
  • XIAO JIANWEI

Assignees

  • 北京合思汇智信息技术有限公司

Dates

Publication Date
20260512
Application Date
20260202

Claims (10)

  1. 1. The rule engine data processing method based on artificial intelligence is characterized by being applied to an enterprise-level automatic data flow system, wherein the enterprise-level automatic data flow system is configured with a plurality of association rules divided according to priority and used for matching and processing multi-dimensional business data of a cross-system, and the method comprises the following steps: responding to completion of matching operation of input data based on a first priority association rule, and inputting first to-be-processed data which is not successfully matched into a pre-trained artificial intelligent model; based on the first data to be processed, predicting by utilizing the artificial intelligent model, and generating an association success rate of the first data to be processed in a second priority association rule; If the association success rate is lower than a preset success rate threshold value, generating a pre-locking instruction for the first data to be processed; based on the pre-lock instruction, the first data to be processed is excluded before a subsequent association rule in a currently executing packet is ready to be executed.
  2. 2. The method according to claim 1, wherein the method further comprises: and when determining the actual execution sequence of each association rule in the current execution group, generating the actual execution sequence by a pre-deployed artificial intelligent scheduling model based on the data volume distribution of the current batch of data to be processed and the system resource load state.
  3. 3. The method according to claim 1, wherein the method further comprises: Responding to a manual execution request initiated by a user, detecting an association rule corresponding to a selected task, and judging whether the association rule with higher execution sequence but not yet executed exists or not; If the user-defined task is in existence, inputting the rule characteristics comprising the association rule corresponding to the selected task, the configuration information of the unexecuted association rule with higher priority and the log data of the history matching success rate into a pre-trained artificial intelligent risk assessment model.
  4. 4. A method according to claim 3, characterized in that the method further comprises: based on the output result of the artificial intelligence risk assessment model, generating a risk assessment report containing a data error proportion predicted value and a downstream business influence level; generating interaction prompt information of corresponding level based on the downstream business influence level in the risk assessment report; and when the downstream business influence level reaches a serious level, a forced confirmation mechanism is started, and the selected task is forbidden to be continuously executed under the condition that the user clear confirmation operation is not received.
  5. 5. The method according to claim 1, wherein the method further comprises: In the association rule configuration stage, responding to the operation of adding or modifying the association rule, calling a pre-trained artificial intelligent recommendation model, and generating a priority suggestion and a priority external association locking strategy suggestion for the rule; The priority suggestion is generated based on matching field combination of rules, filtering strictness, types of related data sources, expected processed data quantity and historical matching success rate, and the priority outer association locking policy suggestion comprises at least one of recommending locking parent-level associated data, locking child-level associated data or locking all associated data.
  6. 6. The method according to claim 1, wherein the method further comprises: And regularly receiving matching execution logs and user feedback data in a preset period, and training the artificial intelligent model based on the user feedback data so as to update parameters of the artificial intelligent model.
  7. 7. The method of claim 6, wherein the method further comprises: recording the deviation between the recommended value output by the artificial intelligent model and the actual configuration value of the user, and inputting the deviation as a negative sample into the next training process of the artificial intelligent model to correct the model deviation.
  8. 8. The rule engine data processing system based on artificial intelligence is characterized by being used for multi-dimensional business data matching processing in an enterprise-level automation data flow system, wherein the enterprise-level automation data flow system is configured with a plurality of association rules divided by priority, and the rule engine data processing system based on artificial intelligence comprises: the data input module is used for responding to completion of matching operation of input data based on a first priority association rule, and first to-be-processed data which is not successfully matched exists, and the first to-be-processed data is input into a pre-trained artificial intelligent model; The model prediction module is used for predicting by utilizing the artificial intelligent model based on the first data to be processed and generating the association success rate of the first data to be processed in a second priority association rule; the pre-locking decision module is used for generating a pre-locking instruction aiming at the first data to be processed if the association success rate is lower than a preset success rate threshold value; and the flow control module is used for eliminating the first data to be processed before the execution of the subsequent association rule in the current execution packet is prepared based on the pre-locking instruction.
  9. 9. An electronic device comprising a memory, a processor, the memory having stored therein a computer program executable on the processor, characterized in that the processor, when executing the computer program, implements the steps of the method of any of the preceding claims 1 to 7.
  10. 10. A computer-readable storage medium, wherein the computer-readable storage medium stores computer-executable instructions, the computer-executable instructions, when invoked and executed by a processor, cause the processor to perform the method of any one of claims 1 to 7.

Description

Rule engine data processing method and system based on artificial intelligence and electronic equipment Technical Field The invention relates to the technical field of data processing, in particular to a rule engine data processing method, system and electronic equipment based on artificial intelligence. Background In the field of automated data processing for enterprise-level information systems, rule engines are widely used to implement associative matching and flow control between multi-source heterogeneous data. Such systems typically screen, compare, and bind mass data based on preset business logic to support critical business processes such as financial reconciliation, order settlement, contract performance, and the like. The traditional rule engine adopts a static configuration mechanism, and a user needs to manually define the execution priority, the matching condition and the data locking strategy of each rule and execute the rules in sequence according to a set sequence. Such methods can meet basic requirements in the face of simple scenes with a small number of rules and stable data structures, but gradually expose the problem of insufficient adaptability in high-concurrency and high-complexity practical applications. As the digitization degree of enterprises increases, data interaction among systems is increasingly frequent, and rule systems show a trend of scale expansion, level complexity and dynamic evolution aggravation. The existing rule engine generally depends on manual experience to complete initial configuration, and is difficult to quickly adjust when facing data distribution change or newly added business scenes. Meanwhile, due to the lack of the ability of prejudging the follow-up behavior of unmatched data, all data need to be tried to be matched step by step, so that a large amount of calculation resources are consumed on an invalid path with low success rate. In addition, in the task execution process, especially in the case of manual intervention operation, the system often cannot effectively identify potential operation risks, and data consistency problems are easily caused. These limitations make conventional rule execution modes a serious challenge in terms of efficiency, accuracy, and security. Disclosure of Invention The invention aims to provide a rule engine data processing method, a rule engine data processing system and electronic equipment based on artificial intelligence, so as to solve the technical problems of low processing efficiency, poor accuracy and low resource utilization rate in the prior art. In a first aspect, an embodiment of the present invention provides a rule engine data processing method based on artificial intelligence, which is applied to an enterprise-level automation data flow system, where the enterprise-level automation data flow system is configured with a plurality of association rules divided by priority and is used for performing matching processing on multi-dimensional business data of a cross system, the method includes, in response to completion of matching operation on input data based on a first priority association rule and presence of unsuccessfully matched first to-be-processed data, inputting the first to-be-processed data to a pre-trained artificial intelligence model, based on the first to-be-processed data, predicting by using the artificial intelligence model, generating an association success rate of the first to-be-processed data in a second priority association rule, if the association success rate is lower than a preset success rate threshold, generating a pre-locking instruction for the first to-be-processed data, and based on the pre-locking instruction, excluding the first to-be-processed data before preparing to execute a subsequent association rule in a current execution packet. In some alternative implementations, the method further includes generating, by a pre-deployed artificial intelligence scheduling model, the actual execution order based on the data size distribution and the system resource load status of the current lot of data to be processed when determining the actual execution order of each association rule within the current execution packet. In some optional implementations, the method further comprises the steps of responding to a manual execution request initiated by a user, detecting an association rule corresponding to the selected task, judging whether an association rule with a higher execution sequence but not yet executed exists, and if so, inputting the rule characteristics comprising the association rule corresponding to the selected task, the configuration information of the non-executed higher priority association rule and log data of history matching success rate into a pre-trained artificial intelligent risk assessment model. In some optional implementations, the method further includes generating a risk assessment report including a data misprediction value and a downstream business impact