Intelligent decision-making method and system based on Drools technology

Abstract

The invention discloses an intelligent decision-making method and system based on a Drools technology, belongs to the technical field of artificial intelligence, and is used for solving the technical problems that in a current tax policy change and application method, application of a business rule engine is shallow, manual auxiliary operation is still needed, and the labor cost cannot be fundamentally saved. The method comprises the following steps: automatically acquiring newly added tax policy information and extracting text information; performing content identification on the extracted text information to obtain a newly added service entity and a newly added service rule; based on the newly added business entity and the newly added business rule, updating the existing entity and the existing rule stored in the rule base; the updated rule base is approved and adjusted through a visual rule adjusting interface; according to the service information of the actual service, extracting a corresponding service rule from a rule base, and generating a corresponding service rule engine through a Drools technology; and through the service rule engine, carrying out service intelligent decision making on the actual service.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an intelligent decision-making method and system based on Drools technology. Background Art

[0002] The current tax policies cover a wide range, are highly refined, and are updated quickly. In order to conduct tax management more efficiently and in compliance, enterprises need to obtain changes more quickly, apply tax policies more comprehensively, and conduct tax control more accurately. The current tax management strategy of enterprises is mainly to obtain policy interpretations through consulting the tax bureau, notifications from superior units, or services provided by third-party consulting companies, and to abstract the interpreted business logic (such as control requirements, applicable preferential policies, etc.) into business rules and feed them back to the information department and business personnel. This model has the following defects:

[0003] 1. After receiving the change request, the information department manually adjusts the business process and then performs downtime deployment after the adjustment. The modification cycle is long and the repetitive cost is high.

[0004] 2. Sales personnel manually judge whether the sales business is applicable to the preferential policies by comparing them with the preferential policies. The workload of sales personnel is heavy and the probability of omissions and errors is high.

[0005] At present, some solutions simply apply the business rule engine Drools to the business process, manually summarize and generalize some rule conditions, and enter them into the system for use. This cannot fundamentally save the labor costs generated by identifying policy changes and policy applications. Summary of the invention

[0006] The embodiment of the present invention provides an intelligent decision-making method and system based on Drools technology, which is used to solve the following technical problems: In the current tax policy change and application methods, the application of business rule engines is relatively shallow, and manual auxiliary operations are still required, which cannot fundamentally save labor costs.

[0007] The embodiment of the present invention adopts the following technical solutions:

[0008] On the one hand, an embodiment of the present invention provides an intelligent decision-making method based on Drools technology, the method comprising: automatically acquiring newly added tax policy information and extracting text information;

[0009] Perform content recognition on the extracted text information to obtain the corresponding new business entities and new business rules;

[0010] Based on the newly added business entities and newly added business rules, the existing entities and existing rules stored in the rule base are updated;

[0011] Approving and adjusting the updated rule base through a visual rule adjustment interface;

[0012] According to the business information of the actual business, the corresponding business rules are extracted from the rule base, and the corresponding business rule engine is generated through Drools technology;

[0013] Through the business rule engine, business intelligent decision-making is performed on the actual business.

[0014] In a feasible implementation manner, before automatically acquiring the newly added tax policy information and extracting text information, the method further includes:

[0015] Deploy a private big model in the enterprise system, and pre-train the big model based on the historical policy information and rule files in the enterprise system to enable it to have logical processing capabilities in tax scenarios;

[0016] Corresponding information acquisition services are added to the enterprise system and knowledge base to obtain issued policies and enterprise internal management policy documents.

[0017] In a feasible implementation, automatically acquiring newly added tax policy information and extracting text information specifically includes:

[0018] Through the information acquisition service, the enterprise system can obtain new tax policy information or newly released internal enterprise management policy documents in real time;

[0019] Determine the information type of the tax policy information or the newly released internal management policy document of the enterprise; wherein the information type includes plain text information and non-text information;

[0020] If the information type is plain text information, directly extract the corresponding text information;

[0021] If the information type is non-text information, the text information therein is extracted by OCR technology.

[0022] In a feasible implementation, content recognition is performed on the extracted text information to obtain the corresponding new business entity and new business rules, which specifically include:

[0023] Performing content recognition on the extracted text information through a natural language processing model to obtain text content information;

[0024] Input the text content information into the large model to identify key information therein; wherein the key information at least includes newly added business entities, entity attributes, keywords, information summaries and context information;

[0025] The logical relationship in the text content information is understood according to the key information, and corresponding new business rules are generated; wherein the new business rules at least include: business verification rules and preferential policy application rules.

[0026] In a feasible implementation, based on the newly added business entity and the newly added business rule, the existing entities and the existing rules stored in the rule base are updated, specifically including:

[0027] Searching the rule base to see whether there is an existing entity identical to the newly added business entity, and if so, comparing the business rules of the existing entity with the newly added business rules, and if there is a change, updating the business rules of the existing entity and generating a change record;

[0028] If there is no existing entity identical to the newly added business entity in the rule base, the newly added business entity and the newly added business rules are stored in the rule base, and a change record is generated.

[0029] In a feasible implementation, the updated rule base is reviewed and adjusted through a visual rule adjustment interface, specifically including:

[0030] When the rule base is changed, the change record is viewed in the visual rule adjustment interface, and corresponding verification rules or approval logic rules are created based on its entity attributes to verify and approve the changed or newly added business entities and business rules;

[0031] When changes are detected in business entities or business rules through manual monitoring, corresponding adjustments are made to existing entities and existing rules in the visual rule adjustment interface.

[0032] In a feasible implementation, according to the business information of the actual business, the corresponding entity rules are extracted from the rule base, and the corresponding business rule engine is generated by Drools technology, which specifically includes:

[0033] Extracting business information from a business list generated by actual business; wherein the business information at least includes an actual business entity corresponding to the actual business;

[0034] It is queried in the rule base whether there is a business rule of the actual business entity. If so, the corresponding business rule is extracted and a corresponding business rule engine is formed through the Drools technology.

[0035] In a feasible implementation manner, the business rule engine is used to make business intelligent decisions on the actual business, specifically including:

[0036] Through the business rule engine, policy application decisions and intelligent compliance audits are performed on the actual business;

[0037] If the policy applicability decision and compliance intelligent audit are all passed, the business documents of the actual business are allowed to be submitted;

[0038] At a preset self-inspection time point or a preset audit time point, if improvements are found in the business document, the corresponding improvement materials are uploaded, the corresponding business rules are obtained again through the big model, and the database is updated or manually adjusted.

[0039] In a feasible implementation, the business rule engine is used to make policy application decisions and conduct intelligent compliance audits on the actual business, specifically including:

[0040] According to the business rules in the business rule engine, determine whether the actual business is applicable to the current tax policy. If not applicable, do not allow the submission of the business document of the actual business, and generate relevant prompt information according to the business rules;

[0041] According to the business entity in the business rule engine, it is determined whether the entity attributes of the business entity of the actual business meet the specification requirements. If not, the submission of the business document of the actual business is not allowed.

[0042] On the other hand, an embodiment of the present invention further provides an intelligent decision-making system based on Drools technology, the system comprising:

[0043] The business rule acquisition module is used to automatically acquire new tax policy information and extract text information; perform content recognition on the extracted text information to obtain the corresponding new business entities and new business rules; based on the new business entities and new business rules, update the existing entities and existing rules stored in the rule library;

[0044] A business rule temporary adjustment module, used to review and adjust the updated rule base through a visual rule adjustment interface;

[0045] The business rule application module is used to extract corresponding business rules from the rule base according to the business information of the actual business, and generate a corresponding business rule engine through Drools technology; through the business rule engine, business intelligent decision-making is performed on the actual business.

[0046] Compared with the prior art, the intelligent decision-making method and system based on Drools technology provided by the embodiment of the present invention has the following beneficial effects:

[0047] The present invention uses artificial intelligence technology to automatically identify control points and preferential points in corporate tax policies, automatically form business rules, and hot deploy them to business processes for application without restarting the service, solving the cost problem of frequent changes and deployment of business rules when tax policies are updated and iterated quickly and the degree of corporate control is different. It greatly shortens the cost and cycle of changing corporate tax verification rules, allowing companies to conduct tax management and apply preferential policies in a comprehensive and compliant manner. With the continuous application of this system, the generation of rules will become more and more accurate, greatly reducing the labor cost caused by rule changes. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments recorded in the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0049] Figure 1 A flow chart of an intelligent decision-making method based on Drools technology provided by an embodiment of the present invention;

[0050] Figure 2 A structural diagram of an intelligent decision-making system based on Drools technology provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to enable those skilled in the art to better understand the technical solutions in the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of this specification, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0052] The embodiment of the present invention provides an intelligent decision-making method based on Drools technology, such as Figure 1 As shown, the intelligent decision-making method based on Drools technology specifically includes steps S101-S106:

[0053] S101. Automatically obtain newly added tax policy information and extract text information.

[0054] Specifically, a privatized large model is deployed in the enterprise system, and based on the historical policy information and rule files in the enterprise system, the large model is pre-trained to enable it to have logical processing capabilities in tax scenarios.

[0055] Furthermore, corresponding information acquisition services are added to the enterprise system and knowledge base to obtain issued policies and internal enterprise management policy documents.

[0056] Furthermore, through the information acquisition service, the enterprise system can obtain the newly received tax policy information or the newly released enterprise internal management policy documents in real time, and determine the information type of the tax policy information or the newly released enterprise internal management policy documents; wherein the information type includes plain text information and non-text information.

[0057] Furthermore, if the information type of the tax policy information or the newly released internal management policy document of the enterprise is plain text information, the corresponding text information is directly extracted; if it is non-text information, the text information is extracted through OCR technology.

[0058] S102: Perform content recognition on the extracted text information to obtain corresponding new business entities and new business rules.

[0059] Specifically, the extracted text information is identified through the natural language processing model to obtain text content information. The text content information is input into the large model to identify key information therein; wherein the key information at least includes newly added business entities, entity attributes, keywords, information summaries and context information.

[0060] Furthermore, the logical relationship in the text content information is understood based on the key information, and corresponding new business rules are generated; wherein the new business rules at least include: business verification rules and preferential policy application rules.

[0061] S103: Based on the newly added business entity and the newly added business rule, the existing entities and the existing rules stored in the rule base are updated.

[0062] Specifically, the rule base is searched to see whether there is an existing entity identical to the newly added business entity. If so, the business rules of the existing entity are compared with the newly added business rules. If there are changes, the business rules of the existing entity are updated and a change record is generated.

[0063] Furthermore, if there is no existing entity identical to the newly added business entity in the rule base, the newly added business entity and the newly added business rule are newly stored in the rule base, and a change record is generated.

[0064] S104. Approve and adjust the updated rule base through the visual rule adjustment interface.

[0065] Specifically, when the rule base changes, the change record is viewed in the visual rule adjustment interface, and based on its entity attributes, corresponding verification rules or approval logic rules are created to verify and approve the changed or newly added business entities and business rules.

[0066] Furthermore, when changes in business entities or business rules are detected manually, corresponding adjustments are made to existing entities and existing rules in the visual rule adjustment interface.

[0067] S105. According to the business information of the actual business, the corresponding business rules are extracted from the rule base, and the corresponding business rule engine is generated through the Drools technology.

[0068] Specifically, the business information is extracted from the business list generated by the actual business; wherein the business information at least includes the actual business entity corresponding to the actual business.

[0069] Furthermore, it is queried in the rule library whether there are business rules for the actual business entity. If so, the corresponding business rules are extracted and a corresponding business rule engine is formed through the Drools technology.

[0070] S106. Make intelligent business decisions on actual business through the business rule engine.

[0071] Specifically, the business rule engine is used to make policy applicability decisions and intelligent compliance audits for actual business. If both policy applicability decisions and intelligent compliance audits are passed, the business documents of the actual business are allowed to be submitted.

[0072] As a feasible implementation method, the business rule engine is used to make policy application decisions and intelligent compliance audits for actual businesses, including:

[0073] Based on the business rules in the business rule engine, determine whether the actual business is applicable to the current tax policy. If not, the business documents of the actual business are not allowed to be submitted, and relevant prompt information is generated according to the business rules. Based on the business entities in the business rule engine, determine whether the entity attributes of the business entities of the actual business meet the regulatory requirements. If not, the business documents of the actual business are not allowed to be submitted.

[0074] Furthermore, at a preset self-inspection time point or a preset audit time point, if improvements are found in the business document, the corresponding improvement materials are uploaded, the corresponding business rules are obtained again through the big model, and the database is updated or manually adjusted.

[0075] As a feasible implementation method, the intelligent decision-making method based on Drools technology provided by the present invention mainly includes the following parts: automatically obtaining business decision rules through an artificial intelligence big model, temporary adjustment of decision rules, application and adjustment of decision rules.

[0076] 1. Automatically obtain decision rules:

[0077] First, a privatized large model is deployed in the enterprise, and the large model is pre-trained based on the company's historical policies and rule documents to enable it to have logical processing capabilities in tax scenarios.

[0078] Then, corresponding information acquisition services are added to the company's OA system and knowledge base to obtain official policies and internal management policy documents or notices. After obtaining the corresponding information subject, first determine its information type. If it is pure text information, extract it directly. If it is non-text information such as PDF or pictures, extract relevant information through OCR technology.

[0079] After reading the text information, the text content is understood through natural language processing technology. Then, based on the trained large model, the corresponding business entity, attributes, keywords, information summary and other key information are identified, and the logical relationship between them is understood according to the context, and finally the rules of the corresponding entity are formed. The rules will involve business verification, preferential policy application rules, etc.

[0080] For example:

[0081] ①Business verification rules: Prohibition category: Reimbursement of invoices from other units is prohibited.

[0082] ② Rules for application of preferential policies: If...then...: If the purchased agricultural products are used to manufacture 13% of the goods, the preferential policy of additional deduction of 1% can be applied.

[0083] Furthermore, the corresponding business rules are stored in the database for dynamic access and updating. Every time a new rule is extracted, it is compared with the rules in the database. If the same entity exists, the rule is compared to see if there is any change. If yes, it is updated, otherwise it is skipped. If the entity rule is completely new, a new rule record is added. All rule additions and updates will record the change records for later tracing and reference.

[0084] 2. Temporary adjustment of rules:

[0085] Due to the influence of large model computing power, the accuracy cannot reach 100%, so this system provides a visual rule adjustment interface, allowing users to create and adjust rules through the visual interface, including:

[0086] ① Add entity rules: Create validation rules or logic rules for entity attributes.

[0087] ② Edit existing entity rules: adjust existing entity rules.

[0088] ③ Change approval: Changes to the rules must be approved by the competent authorities and legal affairs department, and the actual changes will only be completed after approval.

[0089] ④View the change records.

[0090] 3. Application and adjustment of decision-making:

[0091] First of all, all rules can be hot-deployed and take effect immediately through the characteristics of Drools. That is, whenever new policies and management documents are issued internally, if new rules are generated, subsequent business can be applied immediately.

[0092] For the sake of explanation, the application scenario uses the invoicing process of sales personnel selling goods as an example:

[0093] ① After completing the sale, the salesperson will fill out the sales list and enter the product information. After entering the product information, the intelligent decision-making system will query the rule library based on the product entity to see if there is a rule record for the product entity. If so, the corresponding rules will be extracted, and a rule engine will be formed through Drools, which will be cached in the system waiting for a decision.

[0094] ② First, make a decision on the applicability of the policy. According to the corresponding rules, determine whether the current business is applicable to the current policy. If it is not applicable, supplement the relevant information and provide prompts according to the rules.

[0095] ③ Then conduct intelligent compliance audit. According to the rule engine, determine whether the relevant attributes of the current entity meet the regulatory requirements. If not, the business personnel will not be allowed to submit the document. If compliant, submission will be allowed.

[0096] ④After submission, the subsequent business process will continue.

[0097] ⑤ If improvements are found during the self-inspection of business documents at the end of the month or other audit time points, the corresponding improvement materials can be uploaded, and the corresponding rules can be obtained again through the big model for update or manual adjustment.

[0098] ⑥ At the same time, manual adjustments and updates of improvement materials will continue to train the large model and improve the accuracy of subsequent rule extraction of the large model.

[0099] Based on the above logic, as the rules are continuously applied, the generation of rules will become more and more accurate, greatly reducing the labor costs caused by rule changes.

[0100] In addition, the embodiment of the present invention also provides an intelligent decision-making system based on Drools technology, such as Figure 2 As shown, the intelligent decision-making system 200 based on Drools technology specifically includes:

[0101] The business rule acquisition module 210 is used to automatically acquire the newly added tax policy information and extract text information; perform content recognition on the extracted text information to obtain the corresponding newly added business entities and newly added business rules; based on the newly added business entities and newly added business rules, update the existing entities and existing rules stored in the rule library;

[0102] A business rule temporary adjustment module 220 is used to review and adjust the updated rule base through a visual rule adjustment interface;

[0103] The business rule application module 230 is used to extract corresponding business rules from the rule base according to the business information of the actual business, and generate a corresponding business rule engine through the Drools technology; through the business rule engine, business intelligent decision-making is performed on the actual business.

[0104] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0105] This specification may be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform specific tasks or implement specific abstract data types. This specification may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected through a communication network. In a distributed computing environment, program modules may be located in local and remote computer storage media, including storage devices.

[0106] Each embodiment of the present invention is described in a progressive manner, and the same or similar parts between the embodiments can be referred to each other, and each embodiment focuses on the differences from other embodiments. In particular, for the device, equipment, and non-volatile computer storage medium embodiments, since they are basically similar to the method embodiments, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiments.

[0107] The above describes specific embodiments of the present invention. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0108] The above description is only an embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the embodiments of the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the embodiments of the present invention should be included in the protection scope of the present invention.

Claims

1. An intelligent decision-making method based on Drools technology, characterized in that: The method comprises: Automatically obtain new tax policy information and extract text information; Perform content recognition on the extracted text information to obtain the corresponding new business entities and new business rules; Based on the newly added business entities and newly added business rules, the existing entities and existing rules stored in the rule base are updated; Approving and adjusting the updated rule base through a visual rule adjustment interface; According to the business information of the actual business, the corresponding business rules are extracted from the rule base, and the corresponding business rule engine is generated through Drools technology; Through the business rule engine, business intelligent decision-making is performed on the actual business.

2. According to claim 1, the intelligent decision-making method based on Drools technology is characterized in that: Before automatically acquiring the newly added tax policy information and extracting text information, the method further includes: Deploy a private big model in the enterprise system, and pre-train the big model based on the historical policy information and rule files in the enterprise system to enable it to have logical processing capabilities in tax scenarios; Corresponding information acquisition services are added to the enterprise system and knowledge base to obtain issued policies and enterprise internal management policy documents.

3. The intelligent decision-making method based on Drools technology according to claim 2 is characterized in that: Automatically obtain new tax policy information and extract text information, including: Through the information acquisition service, the enterprise system can obtain new tax policy information or newly released internal enterprise management policy documents in real time; Determine the information type of the tax policy information or the newly released internal management policy document of the enterprise; wherein the information type includes plain text information and non-text information; If the information type is plain text information, directly extract the corresponding text information; If the information type is non-text information, the text information therein is extracted by OCR technology.

4. The intelligent decision-making method based on Drools technology according to claim 2 is characterized in that: Perform content recognition on the extracted text information to obtain the corresponding new business entities and new business rules, including: Performing content recognition on the extracted text information through a natural language processing model to obtain text content information; Input the text content information into the large model to identify key information therein; wherein the key information at least includes newly added business entities, entity attributes, keywords, information summaries and context information; The logical relationship in the text content information is understood according to the key information, and corresponding new business rules are generated; wherein the new business rules at least include: business verification rules and preferential policy application rules.

5. The intelligent decision-making method based on Drools technology according to claim 1 is characterized in that: Based on the newly added business entities and new business rules, the existing entities and rules stored in the rule base are updated, including: Searching the rule base to see whether there is an existing entity identical to the newly added business entity, and if so, comparing the business rules of the existing entity with the newly added business rules, and if there is a change, updating the business rules of the existing entity and generating a change record; If there is no existing entity identical to the newly added business entity in the rule base, the newly added business entity and the newly added business rules are stored in the rule base, and a change record is generated.

6. The intelligent decision-making method based on Drools technology according to claim 1 is characterized in that: Through the visual rule adjustment interface, the updated rule base is reviewed and adjusted, including: When the rule base is changed, the change record is viewed in the visual rule adjustment interface, and corresponding verification rules or approval logic rules are created based on its entity attributes to verify and approve the changed or newly added business entities and business rules; When changes are detected in business entities or business rules through manual monitoring, corresponding adjustments are made to existing entities and existing rules in the visual rule adjustment interface.

7. The intelligent decision-making method based on Drools technology according to claim 1 is characterized in that: According to the business information of the actual business, the corresponding entity rules are extracted from the rule base, and the corresponding business rule engine is generated through the Drools technology, specifically including: Extracting business information from a business list generated by actual business; wherein the business information at least includes an actual business entity corresponding to the actual business; It is queried in the rule base whether there is a business rule of the actual business entity. If so, the corresponding business rule is extracted and a corresponding business rule engine is formed through the Drools technology.

8. The intelligent decision-making method based on Drools technology according to claim 1 is characterized in that: Through the business rule engine, business intelligent decision-making is performed on the actual business, specifically including: Through the business rule engine, policy application decisions and intelligent compliance audits are performed on the actual business; If the policy applicability decision and compliance intelligent audit are all passed, the business documents of the actual business are allowed to be submitted; At a preset self-inspection time point or a preset audit time point, if improvements are found in the business document, the corresponding improvement materials are uploaded, the corresponding business rules are obtained again through the big model, and the database is updated or manually adjusted.

9. The intelligent decision-making method based on Drools technology according to claim 8, characterized in that: Through the business rule engine, policy application decisions and intelligent compliance audits are conducted on the actual business, including: According to the business rules in the business rule engine, determine whether the actual business is applicable to the current tax policy. If not applicable, do not allow the submission of the business document of the actual business, and generate relevant prompt information according to the business rules; According to the business entity in the business rule engine, it is determined whether the entity attributes of the business entity of the actual business meet the specification requirements. If not, the submission of the business document of the actual business is not allowed.

10. An intelligent decision-making system based on Drools technology, characterized in that: The system comprises: The business rule acquisition module is used to automatically acquire new tax policy information and extract text information; perform content recognition on the extracted text information to obtain the corresponding new business entities and new business rules; based on the new business entities and new business rules, update the existing entities and existing rules stored in the rule library; A business rule temporary adjustment module, used to review and adjust the updated rule base through a visual rule adjustment interface; The business rule application module is used to extract corresponding business rules from the rule base according to the business information of the actual business, and generate a corresponding business rule engine through Drools technology; through the business rule engine, business intelligent decision-making is performed on the actual business.

Images