Rule adjustment method and device based on recommendation model, equipment and medium

Abstract

The invention relates to the technical field of intelligent decision making, and provides a rule adjustment method based on a recommendation model, and the method comprises the steps: obtaining description information corresponding to a rule engine change point; based on the description information, determining associated information corresponding to the rule engine change point, and determining at least one label corresponding to the rule engine change point according to the associated information; based on a target recommendation model, identifying all tags corresponding to the rule engine change point to obtain at least one rule associated with the rule engine change point; and based on a preset adjustment rule, adjusting all rules associated with the rule engine change point. The risk control rule associated with the rule engine change point is automatically adjusted. The risk control rule associated with the rule engine change point is automatically adjusted, the to-be-adjusted rule is prevented from being missed due to manual evaluation, the rule change efficiency and accuracy are improved, the labor cost is reduced, and the user experience is improved.

Description

Technical Field

[0001] The present invention relates to the fields of intelligent decision-making and financial technology, and in particular to a rule adjustment method, device, computer equipment and computer-readable storage medium based on a recommendation model. Background Art

[0002] As people's living standards improve, insurance business is becoming more and more popular. Insurance underwriting is the prerequisite for insurance underwriting and the first step in the insurance business processing flow. Generally, there are more than 100 risk control rules used for insurance underwriting, involving multiple related parties such as new insurance, renewal, and preservation. Once the rule engine is changed, it is necessary to manually adjust the multiple rules related to the change point. When there are too many change points, more rules need to be manually evaluated and adjusted. Therefore, the current manual rule adjustment is prone to omission of the rules to be adjusted.

[0003] Therefore, how to solve the problem that the current manual rule adjustment easily leads to the omission of the rules to be adjusted has become a technical problem that needs to be solved urgently. Summary of the invention

[0004] The main purpose of the present invention is to provide a rule adjustment method, device, equipment and computer-readable storage medium based on a recommendation model, aiming to solve the technical problem that the current manual rule adjustment easily leads to the omission of the rules to be adjusted.

[0005] To achieve the above object, the present invention provides a rule adjustment method based on a recommendation model, the rule adjustment method comprising the following steps:

[0006] When a rule engine modification instruction is detected, the description information corresponding to the rule engine modification point is obtained;

[0007] Based on the description information, determining association information corresponding to the rule engine change point, and determining at least one tag corresponding to the rule engine change point according to the association information;

[0008] Based on the target recommendation model, all tags corresponding to the rule engine change point are identified to obtain at least one rule associated with the rule engine change point;

[0009] Based on the preset adjustment rules, all rules associated with the rule engine change point are adjusted.

[0010] In addition, to achieve the above-mentioned purpose, the present invention further provides a rule adjustment device based on a recommendation model, and the rule adjustment device based on a recommendation model comprises:

[0011] A description information acquisition module is used to acquire description information corresponding to a rule engine change point when a rule engine change instruction is detected;

[0012] A rule tag determination module, configured to determine, based on the description information, association information corresponding to the rule engine change point, and determine at least one tag corresponding to the rule engine change point according to the association information;

[0013] A rule label identification module, used to identify all labels corresponding to the rule engine change point based on the target recommendation model, and obtain at least one rule associated with the rule engine change point;

[0014] The modification rule adjustment module is used to adjust all rules associated with the modification point of the rule engine based on preset adjustment rules.

[0015] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer device, which includes a processor, a memory, and a rule adjustment program based on a recommendation model stored in the memory and executable by the processor, wherein when the rule adjustment program based on the recommendation model is executed by the processor, the steps of the rule adjustment method based on the recommendation model as described above are implemented.

[0016] In addition, to achieve the above-mentioned purpose, the present invention also provides a computer-readable storage medium, on which a rule adjustment program based on a recommendation model is stored, wherein when the rule adjustment program based on the recommendation model is executed by a processor, the steps of the rule adjustment method based on the recommendation model as described above are implemented.

[0017] The present application provides a rule adjustment method based on a recommendation model. When the modification instruction of the rule engine is detected, the method obtains the description information corresponding to the modification point of the rule engine; based on the description information, determines the association information corresponding to the modification point of the rule engine, and determines at least one label corresponding to the modification point of the rule engine according to the association information; based on the target recommendation model, identifies all labels corresponding to the modification point of the rule engine, and obtains at least one rule associated with the modification point of the rule engine; based on the preset adjustment rule, adjusts all rules associated with the modification point of the rule engine. In the above manner, the present application first determines the label corresponding to the modification point of the rule engine, and then determines the risk control rule associated with the modification point of the rule engine by identifying the label, and automatically adjusts the risk control rule associated with the modification point of the rule engine. The risk control rules associated with the modification point of the rule engine are automatically adjusted to avoid the omission of the rules to be adjusted due to manual evaluation, improve the efficiency and accuracy of rule modification, reduce labor costs, and improve user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 It is a flowchart of a first embodiment of a rule adjustment method based on a recommendation model of the present invention;

[0019] Figure 2 It is a flowchart of a second embodiment of a rule adjustment method based on a recommendation model of the present invention;

[0020] Figure 3 It is a flowchart of a third embodiment of a rule adjustment method based on a recommendation model of the present invention;

[0021] Figure 4 is a schematic diagram of a rule adjustment device based on a recommendation model of the present invention;

[0022] Figure 5 The figure is a schematic diagram of the hardware structure of a computer device involved in an embodiment of the present invention.

[0023] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0024] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0025] The flowcharts shown in the accompanying drawings are only examples and do not necessarily include all the contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may also be decomposed, combined or partially merged, so the actual execution order may change according to actual conditions.

[0026] It should be understood that the terms used in this application specification are only for the purpose of describing specific embodiments and are not intended to limit the application. As used in this application specification and the appended claims, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include plural forms.

[0027] It should be further understood that the term “and / or” used in the specification and appended claims refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0028] The recommendation model-based rule adjustment method involved in the embodiment of the present invention is mainly applied to computer equipment. The recommendation model-based rule adjustment equipment can be a PC, a portable computer, a mobile terminal, or other equipment with display and processing functions.

[0029] In conjunction with the accompanying drawings, some embodiments of the present application are described in detail below. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0030] Reference Figure 1 , Figure 1 It is a flowchart diagram of the first embodiment of the rule adjustment method based on the recommendation model of the present invention.

[0031] like Figure 1 As shown, an embodiment of the present invention provides a rule adjustment method based on a recommendation model, and the rule adjustment method based on a recommendation model includes steps S10 to S40.

[0032] Step S10: when a modification instruction of the rule engine is detected, obtaining description information corresponding to the modification point of the rule engine;

[0033] At present, there are more than 100 risk control rules generally used for insurance underwriting, involving multiple related parties such as new insurance, renewal, and preservation. Once the rule engine is changed, multiple rules related to the change point need to be manually adjusted. When there are too many change points, more rules need to be manually evaluated and adjusted. Therefore, the current manual rule adjustment is prone to omission of the rules to be adjusted.

[0034] This embodiment first determines the label corresponding to the rule engine change point, then determines the risk control rule associated with the rule engine change point by identifying the label, and automatically adjusts the risk control rule associated with the rule engine change point. This realizes the automatic adjustment of the risk control rules associated with the rule engine change point, avoids the omission of the rules to be adjusted due to manual evaluation, improves the efficiency and accuracy of rule changes, reduces labor costs, and improves user experience.

[0035] Specifically, the description information is the description information of the change point uploaded by the user and the change point of the rule engine, and the description information includes the associated parties of the change point. When the rule engine is detected to have been changed, a change instruction of the rule engine is generated. Then, the description information corresponding to the change point is obtained.

[0036] Step S20: determining association information corresponding to the rule engine change point based on the description information, and determining at least one tag corresponding to the rule engine change point according to the association information;

[0037] In this embodiment, the description information is segmented and entity names are extracted, and the associated information corresponding to the change point is obtained based on the processed description information, and the label associated with the change point is determined based on each associated word in the associated information.

[0038] Among them, the related information of the underwriting rules is classified in advance, such as big data, claims, channels, products, renewals, quotation inquiries, policies, preservation, customer domains, human verification, calculated functions, tables in the database, etc., and labels corresponding to each underwriting rule are generated based on the classified related information.

[0039] Specifically, the relationship between rules and labels is one-to-one or one-to-many. For example, a one-to-one relationship: Big data three-party risk control rules only rely on big data, so the label of big data three-party risk control rules is big data; for example, a one-to-many relationship: mutually exclusive rules rely on the mutually exclusive pool of products and also rely on policy query historical valid policies, so the labels of mutually exclusive rules are products and policies.

[0040] Step S30: Based on the target recommendation model, all tags corresponding to the rule engine change point are identified to obtain at least one rule associated with the rule engine change point;

[0041] In this embodiment, the target recommendation model is a logistic regression LR model that adopts a recommendation algorithm. Based on the LR model, all labels corresponding to the change points are identified to determine all rules associated with the change points.

[0042] Specifically, the label defines two categories, including yes and no, which are represented by 1 and 0 respectively, thereby converting the label into a numerical feature vector that can be calculated by the model.

[0043] Furthermore, the target recommendation model is used to identify all tags corresponding to the rule engine change point, and obtaining at least one rule associated with the rule engine change point specifically includes:

[0044] Obtaining feature vectors of all labels corresponding to the rule engine change points, and inputting the feature vectors into the target recommendation model;

[0045] A linear weighted sum is performed on the feature vector based on the target recommendation model to obtain an output vector, and all rules associated with the change point of the rule engine are determined based on the output vector.

[0046] In this embodiment, if a logistic regression LR model is used as the target recommendation model, the input of the model is each label (feature vector), x is the feature vector of the label, and w is the weight of the feature value corresponding to the feature vector of the label, that is, the weight of each rule under the feature value, as shown in the following formula:

[0047]

[0048] The output result of the target recommendation model is obtained by linearly weighted summing the input feature vectors, that is, obtaining all the rules for label association.

[0049] Step S40: Based on the preset adjustment rules, all rules associated with the rule engine change point are adjusted.

[0050] In this embodiment, the preset adjustment rules are the adjustment processes corresponding to each underwriting rule, and all rules associated with the change point of the rule engine are adjusted according to the preset adjustment rules.

[0051] This embodiment provides a rule adjustment method based on a recommendation model. When the method detects a change instruction of the rule engine, it obtains the description information corresponding to the change point of the rule engine; based on the description information, it determines the association information corresponding to the change point of the rule engine, and determines at least one label corresponding to the change point of the rule engine according to the association information; based on the target recommendation model, it identifies all labels corresponding to the change point of the rule engine to obtain at least one rule associated with the change point of the rule engine; based on the preset adjustment rule, it adjusts all rules associated with the change point of the rule engine. In the above manner, the present application first determines the label corresponding to the change point of the rule engine, and then determines the risk control rule associated with the change point of the rule engine by identifying the label, and automatically adjusts the risk control rule associated with the change point of the rule engine. It realizes the automatic adjustment of the risk control rules associated with the change point of the rule engine, avoids the omission of the rules to be adjusted due to manual evaluation, improves the efficiency and accuracy of rule changes, reduces labor costs, and improves user experience.

[0052] Reference Figure 2 , Figure 2 It is a flowchart diagram of the second embodiment of the rule adjustment method based on the recommendation model of the present invention.

[0053] Based on the above Figure 1 In the embodiment shown, in this embodiment, the rule adjustment method based on the recommendation model further includes:

[0054] Step S50: when a verification instruction of a verification target is detected, determining at least one tag corresponding to the verification target, wherein the verification target is a new product, a new commodity or a new channel;

[0055] Step S60: Based on the target recommendation model, identify all tags corresponding to the verification target to obtain at least one rule associated with the verification target;

[0056] Step S70: verify the verification target based on all association rules associated with the verification target to achieve automated verification of the verification target.

[0057] At present, when a company has a large business volume and often needs to introduce new products, it needs to repeatedly invest manpower to verify the underwriting rules for new products in accordance with the underwriting rules.

[0058] In this embodiment, in order to support the adjustment of risk control rule functions, all related rules are intelligently recommended according to tags and automatic verification is performed, and all corresponding risk control rules are automatically verified when new products / new products / new channels are connected.

[0059] Specifically, when accessing new products / new commodities / new channels, the verification target can be classified according to labels to determine the feature vectors of the corresponding labels. The feature vectors of the labels are then entered into the target recommendation model to identify and list the underwriting rules associated with the verification target through the recommendation algorithm of the logistic regression model. The identified underwriting rules are then configured (i.e. bound) to the product in accordance with the requirements of each rule, and the commodity corresponding to the product is entered through the front end (i.e. the product is bound to the commodity), the underwriting rules are executed on the commodity, and the rule execution results are checked. This completes the risk control verification of the commodity.

[0060] Further, the verifying the verification target based on all the association rules associated with the verification target specifically includes:

[0061] Based on the preset rule requirements, obtain relevant data corresponding to all association rules of the verification target;

[0062] The verification target is verified based on a preset standard input parameter message format, interface input parameter rules corresponding to all association rules of the verification target, and the relevant data.

[0063] In this embodiment, according to the rule requirements, the key data required by all the association rules of the verification target are obtained as the relevant data corresponding to all the association rules of the verification target. Specifically, data can be obtained from fixed customer five-item type, insert table number type, link number type and other types of data.

[0064] Among them, customer five-item data requires providing five pieces of customer information, namely name, gender, birthday, ID type, ID number, etc. Insert table data refers to the table name of known data reading, and directly reading the data in the table in the database. Link data requires data preparation from the source according to the scenario.

[0065] Then, according to the preset standard input parameter message format and the interface input parameter rules (i.e., interface input parameter requirements) corresponding to all the association rules of the verification target, the above-mentioned related data is used to dynamically transmit the corresponding message, thereby verifying the verification target.

[0066] Furthermore, before identifying all tags corresponding to the verification target based on the target recommendation model and obtaining at least one rule associated with the verification target, the method further includes:

[0067] Obtaining a sales plan corresponding to the verification target, and obtaining product information, product plans, and insurance plans based on the sales plan;

[0068] Based on the product information, product plan and insurance plan, obtain a list of standard rules corresponding to the verification target;

[0069] Based on the standard rule list and all association rules associated with the verification target, the target recommendation model is model verified.

[0070] In this embodiment, a standard rule for the verification target may be determined based on the sales plan corresponding to the verification target, so as to perform a model test on the target recommendation model before determining a label of the verification target according to the target recommendation model according to the standard rule.

[0071] Specifically, first obtain the basic data of the verification target (i.e., new product / new commodity / new channel), such as querying the sales plan through the commodity, obtaining product information, product plan, insurance plan and other information; then obtain the commodity rule list through product information, product plan, insurance plan and other information as the standard rule list corresponding to the verification target.

[0072] Before model verification, it is necessary to prepare the necessary data for verification. The necessary data includes a variety of scenarios, such as policyholder rule hits, insured person rule hits, policyholder rule misses, insured person rule misses, etc. Then, according to the rule requirements, obtain the key data required for all the associated rules of the verification target. Specifically, data can be obtained from fixed customer five-item type, table insertion type, link number type and other types of data. According to the pre-set standard input parameter message format, and the interface input parameter rules (i.e., interface input parameter requirements) corresponding to all the associated rules of the verification target, the above-mentioned related data is used to dynamically transfer parameters to the corresponding message, thereby verifying the verification target. After all rules are executed, the assertion results, expected results and actual results of each rule are recorded, and a front-end page display is provided for testers to check. The model verification of the target recommendation model is thus completed.

[0073] Through the above method, this embodiment provides a rule adjustment method based on the recommendation model. This method realizes the automatic adjustment of risk control rules associated with the change points of the rule engine, avoids the omission of rules to be adjusted due to manual evaluation, improves the efficiency and accuracy of rule changes, reduces labor costs, and improves user experience. In addition, when new products are connected, the risk control rules of the products can be automatically verified. Users only need to check whether the rule execution results meet the requirements, so that testers are freed from repeated and complicated rule verification, thereby improving the testing efficiency of the products, improving the accuracy of the test results of the products, and reducing the labor cost of testing.

[0074] Reference Figure 3 , Figure 3 It is a flowchart diagram of the third embodiment of the rule adjustment method based on the recommendation model of the present invention.

[0075] Based on the above Figure 1 In the embodiment shown, in this embodiment, the step S20 specifically further includes:

[0076] Step S21, determining the associated information corresponding to the rule engine change point based on the description information;

[0077] Step S22, matching the association information corresponding to the rule engine change point with a preset tag library;

[0078] Step S23, when there is no tag matching the associated information corresponding to the rule engine change point in the preset tag library, a reminder message of tags to be supplemented is generated, so that the user can supplement the tags;

[0079] Step S24, when there is a tag matching the associated information corresponding to the rule engine change point in the preset tag library, the tag matching the associated information corresponding to the rule engine change point is determined as the tag corresponding to the rule engine change point determined by the associated information.

[0080] In this embodiment, the associated information of the underwriting rules is classified in advance, such as big data, claims, channels, products, renewals, quotation inquiries, policies, preservation, customer domains, human verification, calculated functions, tables in databases, etc., and labels corresponding to each underwriting rule are generated according to the classified associated information. Labels can be words, phrases or sentences (short or long sentences).

[0081] Then, the description information corresponding to the rule engine change point is segmented and keywords are extracted to obtain the associated information corresponding to the rule engine change point.

[0082] Each character, word or sentence in the associated information is compared with each tag in a preset tag library to determine whether there is a tag in the tag library that matches the associated information.

[0083] When there is a tag matching the associated information corresponding to the rule engine change point in the preset tag library, the tag matching the associated information corresponding to the rule engine change point is determined as the tag corresponding to the rule engine change point determined by the associated information.

[0084] When there is no tag matching the associated information corresponding to the rule engine change point in the preset tag library, a reminder message is generated that the tag needs to be supplemented. The reminder message is displayed through the current interface, and a manual tag addition input box is displayed through the current interface, so that the user can manually add a corresponding tag to the rule engine change point based on the reminder message.

[0085] After receiving the tag manually input by the user, the input tag is associated with the rule engine change point and its corresponding associated information and stored to update and supplement the tag library.

[0086] Furthermore, before determining the association information corresponding to the rule engine change point based on the description information and determining at least one label corresponding to the rule engine change point according to the association information, the method further includes:

[0087] Obtaining the associated information corresponding to all underwriting rules, and performing word segmentation and entity name extraction on the associated information corresponding to all underwriting rules;

[0088] The entity names corresponding to all the extracted underwriting rules are classified, and the classification results are stored as corresponding labels.

[0089] In this embodiment, the associated information of the underwriting rules is classified in advance, such as big data, claims, channels, products, renewals, quotation inquiries, policies, preservation, customer domains, human verification, calculated functions, tables in the database, etc., and labels corresponding to each underwriting rule are generated based on the classified associated information.

[0090] See also Figure 4 , Figure 4 It is a functional module diagram of a rule adjustment device based on a recommendation model provided in an embodiment of the present application.

[0091] like Figure 4 As shown, the rule adjustment device 500 based on the recommendation model includes:

[0092] A description information acquisition module 10 is used to acquire description information corresponding to a rule engine change point when a rule engine change instruction is detected;

[0093] A rule tag determination module 20, configured to determine association information corresponding to the rule engine change point based on the description information, and determine at least one tag corresponding to the rule engine change point according to the association information;

[0094] A rule tag identification module 30, for identifying all tags corresponding to the rule engine change point based on a target recommendation model, and obtaining at least one rule associated with the rule engine change point;

[0095] The modified rule adjustment module 40 is used to adjust all rules associated with the modified point of the rule engine based on preset adjustment rules.

[0096] Furthermore, the rule adjustment device 500 based on the recommendation model further includes a target risk control verification module, and the target risk control verification module further includes:

[0097] a target tag determination unit, configured to determine at least one tag corresponding to a verification target when a verification instruction of a verification target is detected, wherein the verification target is a new product, a new commodity or a new channel;

[0098] an association rule determination unit, configured to identify all tags corresponding to the verification target based on the target recommendation model, and obtain at least one rule associated with the verification target;

[0099] The target risk control verification unit is used to verify the verification target based on all association rules associated with the verification target to achieve automatic verification of the verification target.

[0100] Furthermore, the target risk control verification unit specifically includes:

[0101] A related data acquisition subunit, used to acquire related data corresponding to all association rules of the verification target based on preset rule requirements;

[0102] The target risk control verification subunit is used to verify the verification target based on a preset standard input parameter message format, the interface input parameter rules corresponding to all the association rules of the verification target, and the relevant data.

[0103] Furthermore, the target risk control verification module also includes:

[0104] A sales plan acquisition unit, used to acquire the sales plan corresponding to the verification target, and acquire product information, product plan and insurance plan based on the sales plan;

[0105] A standard rule list acquisition unit, configured to acquire a standard rule list corresponding to the verification target based on the product information, product plan and insurance plan;

[0106] The recommendation model verification unit is used to perform model verification on the target recommendation model based on the standard rule list and all association rules associated with the verification target.

[0107] Furthermore, the rule label identification module specifically includes:

[0108] A feature vector input unit, used to obtain feature vectors of all labels corresponding to the rule engine change points, and input the feature vectors into the target recommendation model;

[0109] An association rule output unit is used to perform linear weighted summation on the feature vector based on the target recommendation model to obtain an output vector, and determine all rules associated with the rule engine change point based on the output vector.

[0110] Furthermore, the rule label determination module specifically includes:

[0111] A correlation information determination unit, configured to determine, based on the description information, the correlation information corresponding to the rule engine change point;

[0112] An information tag matching unit, used to match the associated information corresponding to the rule engine change point with a preset tag library;

[0113] A rule label supplement unit, configured to generate a reminder message indicating that labels are to be supplemented when there is no label matching the associated information corresponding to the rule engine change point in the preset label library, so that the user can supplement the labels;

[0114] A rule tag determination unit is used to determine the tag corresponding to the rule engine change point as the tag corresponding to the rule engine change point determined by the association information when there is a tag matching the association information corresponding to the rule engine change point in the preset tag library.

[0115] Furthermore, the rule adjustment device based on the recommendation model further includes a rule label generation module, and the rule label generation module specifically includes:

[0116] A label name extraction unit, used to obtain the associated information corresponding to all underwriting rules, and perform word segmentation and entity name extraction on the associated information corresponding to all underwriting rules;

[0117] A rule label generation unit is used to classify the entity names corresponding to all the extracted underwriting rules and store the classification results as corresponding labels.

[0118] It should be noted that those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working process of the above-described device and each module can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0119] The above-mentioned device can be implemented in the form of a computer program. Figure 5 Runs on the computer device shown.

[0120] See also Figure 5 , Figure 51 is a schematic block diagram of a computer device provided in an embodiment of the present application. The computer device may be a server.

[0121] See also Figure 5 The computer device includes a processor, a memory and a network interface connected through a system bus, wherein the memory may include a non-volatile storage medium and an internal memory.

[0122] The non-volatile storage medium can store an operating system and a computer program. The computer program includes program instructions, and when the program instructions are executed, the processor can execute any rule adjustment method based on the recommendation model.

[0123] The processor is used to provide computing and control capabilities and support the operation of the entire computer equipment.

[0124] The internal memory provides an environment for the operation of the computer program in the non-volatile storage medium. When the computer program is executed by the processor, the processor can execute any rule adjustment method based on the recommendation model.

[0125] The network interface is used for network communication, such as sending assigned tasks, etc. Those skilled in the art will understand that Figure 5 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0126] It should be understood that the processor may be a central processing unit (CPU), and the processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. Among them, the general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0127] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0128] When a rule engine modification instruction is detected, the description information corresponding to the rule engine modification point is obtained;

[0129] Based on the description information, determining association information corresponding to the rule engine change point, and determining at least one tag corresponding to the rule engine change point according to the association information;

[0130] Based on the target recommendation model, all tags corresponding to the rule engine change point are identified to obtain at least one rule associated with the rule engine change point;

[0131] Based on the preset adjustment rules, all rules associated with the rule engine change point are adjusted.

[0132] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0133] When a verification instruction of a verification target is detected, determining at least one tag corresponding to the verification target, wherein the verification target is a new product, a new product or a new channel;

[0134] Based on the target recommendation model, identifying all tags corresponding to the verification target, and obtaining at least one rule associated with the verification target;

[0135] Based on all the association rules associated with the verification target, the verification target is verified to achieve automated verification of the verification target.

[0136] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0137] Based on the preset rule requirements, obtain relevant data corresponding to all association rules of the verification target;

[0138] The verification target is verified based on a preset standard input parameter message format, interface input parameter rules corresponding to all association rules of the verification target, and the relevant data.

[0139] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0140] Obtaining a sales plan corresponding to the verification target, and obtaining product information, product plans, and insurance plans based on the sales plan;

[0141] Based on the product information, product plan and insurance plan, obtain a list of standard rules corresponding to the verification target;

[0142] Based on the standard rule list and all association rules associated with the verification target, the target recommendation model is model verified.

[0143] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0144] Obtaining feature vectors of all labels corresponding to the rule engine change points, and inputting the feature vectors into the target recommendation model;

[0145] A linear weighted sum is performed on the feature vector based on the target recommendation model to obtain an output vector, and all rules associated with the change point of the rule engine are determined based on the output vector.

[0146] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0147] Based on the description information, determining the associated information corresponding to the rule engine change point;

[0148] Matching the associated information corresponding to the rule engine change point with a preset tag library;

[0149] When there is no tag matching the associated information corresponding to the rule engine change point in the preset tag library, a reminder message of tags to be supplemented is generated to allow the user to supplement the tags;

[0150] When there is a tag matching the associated information corresponding to the rule engine change point in the preset tag library, the tag matching the associated information corresponding to the rule engine change point is determined as the tag corresponding to the rule engine change point determined by the associated information.

[0151] In one embodiment, the processor is used to run a computer program stored in the memory to implement the following steps:

[0152] Obtaining the associated information corresponding to all underwriting rules, and performing word segmentation and entity name extraction on the associated information corresponding to all underwriting rules;

[0153] The entity names corresponding to all the extracted underwriting rules are classified, and the classification results are stored as corresponding labels.

[0154] A computer-readable storage medium is also provided in an embodiment of the present application, wherein the computer-readable storage medium stores a computer program, wherein the computer program includes program instructions, and the processor executes the program instructions to implement any one of the recommendation model-based rule adjustment methods provided in the embodiments of the present application.

[0155] The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, such as a hard disk or memory of the computer device. The computer-readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc., equipped on the computer device.

[0156] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of various equivalent modifications or replacements within the technical scope disclosed in the present application, and these modifications or replacements should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A rule adjustment method based on a recommendation model, characterized in that: The rule adjustment method comprises the following steps: When a rule engine modification instruction is detected, the description information corresponding to the rule engine modification point is obtained; Based on the description information, determining association information corresponding to the rule engine change point, and determining at least one tag corresponding to the rule engine change point according to the association information; Based on the target recommendation model, all tags corresponding to the rule engine change point are identified to obtain at least one rule associated with the rule engine change point; Based on the preset adjustment rules, all rules associated with the rule engine change point are adjusted.

2. The rule adjustment method based on the recommendation model according to claim 1, characterized in that: The rule adjustment method based on the recommendation model also includes: When a verification instruction of a verification target is detected, determining at least one tag corresponding to the verification target, wherein the verification target is a new product, a new product or a new channel; Based on the target recommendation model, identifying all tags corresponding to the verification target, and obtaining at least one rule associated with the verification target; Based on all the association rules associated with the verification target, the verification target is verified to achieve automated verification of the verification target.

3. The rule adjustment method based on the recommendation model according to claim 2, characterized in that: The verifying the verification target based on all the association rules associated with the verification target specifically includes: Based on the preset rule requirements, obtain relevant data corresponding to all association rules of the verification target; The verification target is verified based on a preset standard input parameter message format, interface input parameter rules corresponding to all association rules of the verification target, and the relevant data.

4. The rule adjustment method based on the recommendation model according to claim 2, characterized in that: After identifying all tags corresponding to the verification target based on the target recommendation model and obtaining at least one rule associated with the verification target, the method further includes: Obtaining a sales plan corresponding to the verification target, and obtaining product information, product plans, and insurance plans based on the sales plan; Based on the product information, product plan and insurance plan, obtain a list of standard rules corresponding to the verification target; Based on the standard rule list and all association rules associated with the verification target, the target recommendation model is model-verified.

5. The rule adjustment method based on the recommendation model according to claim 1, characterized in that: The target recommendation model-based, identifying all tags corresponding to the rule engine change point, and obtaining at least one rule associated with the rule engine change point specifically includes: Obtaining feature vectors of all labels corresponding to the rule engine change points, and inputting the feature vectors into the target recommendation model; A linear weighted sum is performed on the feature vector based on the target recommendation model to obtain an output vector, and all rules associated with the change point of the rule engine are determined based on the output vector.

6. The rule adjustment method based on the recommendation model according to claim 1, characterized in that: The determining, based on the description information, the association information corresponding to the rule engine change point, and determining, according to the association information, at least one tag corresponding to the rule engine change point specifically includes: Based on the description information, determining the associated information corresponding to the rule engine change point; Matching the associated information corresponding to the rule engine change point with a preset tag library; When there is no tag matching the associated information corresponding to the rule engine change point in the preset tag library, a reminder message of tags to be supplemented is generated to allow the user to supplement the tags; When there is a tag matching the associated information corresponding to the rule engine change point in the preset tag library, the tag matching the associated information corresponding to the rule engine change point is determined as the tag corresponding to the rule engine change point determined by the associated information.

7. The rule adjustment method based on the recommendation model according to any one of claims 1 to 6, characterized in that: Before determining the association information corresponding to the rule engine change point based on the description information, and determining at least one label corresponding to the rule engine change point according to the association information, the method further includes: Obtaining the associated information corresponding to all underwriting rules, and performing word segmentation and entity name extraction on the associated information corresponding to all underwriting rules; The entity names corresponding to all the extracted underwriting rules are classified, and the classification results are stored as corresponding labels.

8. A rule adjustment device based on a recommendation model, characterized in that: The rule adjustment device based on the recommendation model includes: A description information acquisition module is used to acquire description information corresponding to a rule engine change point when a rule engine change instruction is detected; A rule tag determination module, configured to determine, based on the description information, association information corresponding to the rule engine change point, and determine at least one tag corresponding to the rule engine change point according to the association information; A rule label identification module, used to identify all labels corresponding to the rule engine change point based on the target recommendation model, and obtain at least one rule associated with the rule engine change point; The modification rule adjustment module is used to adjust all rules associated with the modification point of the rule engine based on preset adjustment rules.

9. A computer device, characterized in that: The computer device includes a processor, a memory, and a rule adjustment program based on a recommendation model stored in the memory and executable by the processor, wherein when the rule adjustment program based on the recommendation model is executed by the processor, the steps of the rule adjustment method based on the recommendation model as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a rule adjustment program based on a recommendation model, wherein when the rule adjustment program based on a recommendation model is executed by a processor, the steps of the rule adjustment method based on a recommendation model as claimed in any one of claims 1 to 7 are implemented.

Images