Model-based rule strategy rehearsal method and device, computer equipment and medium
By introducing a model-based rule strategy rehearsal method, the problems of inefficiency and increased risks caused by relying on manual assessment in the traditional insurance operation model are solved, and the quantitative evaluation and optimization of strategies and rules are achieved.
Patent Information
- Application Number
- CN202411487335.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-23
- Publication Date
- 2025-02-18
AI Technical Summary
Traditional insurance operation models rely too much on manual review and intuitive empirical judgment when evaluating the effectiveness of strategies and rules, and lack a standardized quantitative evaluation system, resulting in inefficiency and increased operational risks.
A model-based rule strategy rehearsal method is introduced, and quantitative evaluation is carried out through a preset rule strategy rehearsal model, including a preset rehearsal model and an index evaluation efficiency model, to generate a rehearsal success rate and total score.
It improves the efficiency of quantitative evaluation of strategies and rules, reduces human participation, saves time and costs, reduces operational risks, and provides a scientific basis for strategy optimization.
Smart Images
Figure CN119476988A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the fields of intelligent decision-making technology and financial technology, and in particular to a model-based rule strategy preview method, device, computer equipment and medium. Background Art
[0002] In the insurance and finance sector, it is crucial for insurance companies to build a systematic strategy and rules framework to drive business operations and growth. Ensuring the accurate planning and efficient implementation of these operational strategies is the core of maintaining market competitive advantage. However, the traditional insurance operating model relies too much on manual review and intuitive judgment when evaluating the effectiveness of strategies and rules, and lacks a standardized quantitative evaluation system. This model not only significantly reduces work efficiency, but also relies on cumbersome manual evaluation processes when deciding whether certain strategies or rules need to be revoked or adjusted, which is time-consuming and laborious. More importantly, this method makes it difficult to fully examine all potential scenarios and their possible consequences, thereby invisibly increasing the risk level in the operation process.
[0003] Therefore, those skilled in the art are in urgent need of finding a new technical solution to solve the above problems. Summary of the invention
[0004] Based on this, a model-based rule strategy preview method, device, computer equipment and medium are provided to solve the technical problem that the existing technology relies on manual evaluation of strategies and rules, which is time-consuming, labor-intensive and has operational risks.
[0005] A model-based rule strategy preview method, comprising: Obtain data for previewing the rule strategy; After the data to be previewed for the rule strategy are input into the preset rule strategy preview model, the preview success rate and the total score of the data to be previewed for the rule strategy are obtained; the preset rule strategy preview model includes a preset preview model and a preset indicator evaluation effectiveness model; the processing method of the preset rule strategy preview model includes, after the data to be previewed for the rule strategy are input into the preset rule strategy preview model, obtaining the preview success rate corresponding to the data to be previewed for the rule strategy and at least one expected evaluation indicator corresponding to the preview success rate through different rule strategy parameters in the preset rule strategy preview model; the processing method of the preset indicator evaluation effectiveness model includes obtaining at least one total score corresponding to the data to be previewed for the rule strategy through the expected evaluation indicators in the preset indicator evaluation effectiveness model, and the expected evaluation indicators affect the model evaluation indicators in the preset indicator evaluation effectiveness model.
[0006] A model-based rule strategy preview device, comprising: An acquisition module is used to acquire data to be previewed for rule strategies; An input module is used to input the data to be previewed of the rule strategy into a preset rule strategy preview model, and then obtain the preview success rate and total score of the data to be previewed of the rule strategy; the preset rule strategy preview model includes a preset preview model and a preset indicator evaluation effectiveness model; the processing method of the preset rule strategy preview model includes, after the data to be previewed of the rule strategy is input into the preset rule strategy preview model, obtaining the preview success rate corresponding to the data to be previewed of the rule strategy and at least one expected evaluation indicator corresponding to the preview success rate through different rule strategy parameters in the preset rule strategy preview model; the processing method of the preset indicator evaluation effectiveness model includes obtaining at least one total score corresponding to the data to be previewed of the rule strategy through the expected evaluation indicators in the preset indicator evaluation effectiveness model, and the expected evaluation indicators affect the model evaluation indicators in the preset indicator evaluation effectiveness model.
[0007] A computer device comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned model-based rule strategy preview method when executing the computer program.
[0008] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the model-based rule strategy preview method is implemented.
[0009] The above-mentioned model-based rule strategy preview method, device, computer equipment and medium obtain data to be previewed for the rule strategy; After the data to be previewed for the rule strategy are input into the preset rule strategy preview model, the preview success rate and the total score of the data to be previewed for the rule strategy are obtained; the preset rule strategy preview model includes a preset preview model and a preset indicator evaluation effectiveness model; the processing method of the preset rule strategy preview model includes, after the data to be previewed for the rule strategy are input into the preset rule strategy preview model, obtaining the preview success rate corresponding to the data to be previewed for the rule strategy and at least one expected evaluation indicator corresponding to the preview success rate through different rule strategy parameters in the preset rule strategy preview model; the processing method of the preset indicator evaluation effectiveness model includes obtaining at least one total score corresponding to the data to be previewed for the rule strategy through the expected evaluation indicators in the preset indicator evaluation effectiveness model, and the expected evaluation indicators affect the model evaluation indicators in the preset indicator evaluation effectiveness model. This technical solution introduces an intelligent preset rule strategy preview model to replace the original manual processing method. The model also specifically sets up a preset preview model for predicting the success rate and a preset indicator evaluation efficiency model for scoring. The model can quantitatively evaluate strategies and rules, which can improve the efficiency of decision-making or evaluation. Decision-making or evaluation does not require human participation, which can save time and cost. At the same time, the evaluation process of the model can comprehensively evaluate the impact of multiple indicators in the operation process, provide a scientific basis for strategy optimization, effectively reduce operational risks, and ensure that enterprises can respond to market changes more accurately and efficiently, continuously improve their competitiveness, and inject new vitality into the healthy development of the industry. BRIEF DESCRIPTION OF THE DRAWINGS
[0010] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments of the present invention will be briefly introduced below. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative labor.
[0011] Figure 1 It is a schematic diagram of an application environment of a model-based rule strategy preview method in one embodiment of the present invention; Figure 2 is a process diagram of a model-based rule strategy preview method in one embodiment of the present invention; Figure 3 is a schematic structural diagram of a model-based rule strategy preview device in one embodiment of the present invention; Figure 4 is a schematic diagram of a computer device in one embodiment of the present invention. DETAILED DESCRIPTION
[0012] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0013] The model-based rule strategy preview method provided by the present invention can be applied in Figure 1 In an application environment, a client communicates with a server via a network. The client may be, but is not limited to, various personal computers, laptops, smart phones, tablet computers, and portable wearable devices. The server may be implemented as an independent server or a server cluster consisting of multiple servers.
[0014] In one embodiment, if Figure 2 As shown, a model-based rule strategy preview method is provided, which is applied in Figure 1 The server in the example is used as an example to illustrate the process, including the following steps: S10, obtaining data for previewing the rule strategy.
[0015] It can be understood that the data to be used for rule strategy preview is the historical operational data before and after the implementation of the rule strategy (the data to be used for rule strategy preview may include at least one rule strategy), which includes but is not limited to business data, market data and customer feedback data, and the above-mentioned data is not only applicable to insurance and financial scenarios, but also to related financial scenarios, such as banking, securities and other scenarios.
[0016] S20, after inputting the data to be previewed of the rule strategy into the preset rule strategy preview model, obtaining the preview success rate and total score of the data to be previewed of the rule strategy; the preset rule strategy preview model includes a preset preview model and a preset indicator evaluation effectiveness model; the processing method of the preset rule strategy preview model includes, after inputting the data to be previewed of the rule strategy into the preset rule strategy preview model, obtaining the preview success rate corresponding to the data to be previewed of the rule strategy and at least one expected evaluation indicator corresponding to the preview success rate through different rule strategy parameters in the preset rule strategy preview model; the processing method of the preset indicator evaluation effectiveness model includes obtaining at least one total score corresponding to the data to be previewed of the rule strategy through the expected evaluation indicators in the preset indicator evaluation effectiveness model, and the expected evaluation indicators affect the model evaluation indicators in the preset indicator evaluation effectiveness model.
[0017] Understandably, the preset rule strategy preview model is a large data model, which includes a preset preview model and a preset indicator evaluation effectiveness model. The functions corresponding to each model are inconsistent. The preset preview model can predict the performance of the strategy in the actual scenario before the rule strategy goes online, effectively identify potential limitations and problems, provide opportunities for strategy optimization, and avoid the occurrence of potential problems. It can generate a prediction success rate and at least one expected evaluation indicator that can affect the model evaluation indicator in the preset indicator evaluation effectiveness model (specifically, after the new rule strategy parameter encoding is completed, it is input into the trained model and the rule strategy is output). By establishing an operational indicator evaluation model, the effect of the rule strategy is converted into a specific quantitative data indicator - the total score, where the total score can be at least one (for example, one rule strategy corresponds to one total score), which can more accurately measure the actual value of the strategy, provide data support for subsequent decision-making, ensure the accurate implementation and continuous optimization of the strategy, and at the same time, the unified rule effectiveness evaluation can achieve the "slimming" of the rule strategy, save manpower for offline review of the rule, and improve the accuracy of the rule strategy; This embodiment is a processing process of a preset rule strategy preview model. Before using the model for processing, the model can be pre-trained. Only when the parameters of the model meet the specified model parameters can the structure of the model be used for processing.
[0018] It should be noted that the preset rule strategy preview model mentioned here, as well as the preset preview model and the preset indicator evaluation performance model can be one or more, and different models can handle different rule strategies, such as one model processes a rule strategy under one rule strategy parameter.
[0019] In the embodiment of step S10 to step S20, an intelligent preset rule strategy preview model is introduced to replace the original manual processing method, wherein the model also specifically sets up a preset preview model for predicting the success rate and a preset indicator evaluation efficiency model for scoring. The model can quantitatively evaluate strategies and rules, which can improve the efficiency of decision-making or evaluation. Decision-making or evaluation does not require human participation, which can save time and cost. At the same time, the evaluation process of the model can comprehensively evaluate the impact of multiple indicators in the operation process, provide a scientific basis for strategy optimization, effectively reduce operational risks, and ensure that enterprises can respond to market changes more accurately and efficiently, continuously improve their competitiveness, and inject new vitality into the healthy development of the industry.
[0020] Furthermore, after inputting the data to be previewed for the rule strategy into a preset rule strategy preview model, before obtaining the preview success rate and total score of the data to be previewed for the rule strategy, it also includes: Collect historical data related to the rule strategy preview to be conducted; The historical data is preprocessed to obtain training data with outliers removed and / or value filling completed.
[0021] Understandably, historical data includes customer information, business operation data (number of WeChat replies, car binding rate, add-on rate, registration rate, number of APP active times, number of touches, etc.), market data, conversion results, etc. These data should include conversion situations under different strategies. Historical data is data used for training, and the data to be previewed for rule strategies mentioned above are data under different usage scenarios; data preprocessing includes data cleaning, denoising and normalization. Data cleaning is the process of re-examining and verifying data, with the purpose of deleting duplicate information, correcting existing errors, and providing data consistency. This embodiment specifically removes outliers in the data, and fills missing values with the mode and mean to ensure data quality; This embodiment can significantly improve the data quality and analysis accuracy of training data through data preprocessing.
[0022] Furthermore, before performing data preprocessing on the historical data to obtain training data with outliers removed and / or value filling completed, the process further includes: Encoding the rule strategy parameters corresponding to the historical data to obtain encoded training data; Or / and, performing data conversion on preset type data corresponding to the historical data to obtain training data after data conversion.
[0023] Understandably, encoding is a rule strategy encoding. Specifically, the rule strategy parameters are converted into features that can be understood by the model (numeric or categorical features). For example, the delivered services can be converted into the total value of the services (numeric), and the promotional activities can be converted into the activity types (categorical). More specifically, in the process of converting the rule strategy into a feature vector that can be understood by the model, different encoding methods can be used for different data types. For example, the user behavior frequency threshold can be standardized or normalized, and the user type can be encoded using one-hot encoding. Data conversion is a process of converting data of a specific data type into data of a target data type. Specifically, one-hot encoding is used to convert categorical data (values are categories, and there is no order relationship between categories) into numerical data, and label encoding is used to convert ordinal data (values are categories, and there is an order relationship between categories) into numerical data. The encoded rule strategy parameters of this embodiment can be directly input into the machine learning model without additional conversion or interpretation steps; through encoding, the model can better understand and utilize the information in the rule strategy parameters, thereby improving the accuracy of prediction; the encoding process can be automated, reducing manual intervention and improving the efficiency and consistency of data processing; through conversion, inconsistencies, missing values and other problems in the data can be eliminated, and the integrity and accuracy of the data can be improved; the converted data is more in line with the input requirements of the machine learning model, which helps the model to better learn the characteristics and laws of the data; through appropriate conversion, the model can be better adapted to different types of data, thereby improving the generalization ability and robustness of the model.
[0024] It should be noted that the training data after removing outliers and / or completing value filling, the encoded training data and the training data after data conversion can be used as the total training data or as training data separately.
[0025] Furthermore, before obtaining the preview success rate corresponding to the data to be previewed for the rule strategy and at least one expected evaluation index corresponding to the preview success rate through different rule strategy parameters in the preset rule strategy preview model, the method further includes: Filter out target features from the training data according to a preset data filtering method; The target features are input into a preset model structure, and the preset model structure is trained using training data with annotated success rate labels. After confirming that the model parameters adjusted by a preset model parameter adjustment method meet the preset model parameters, the preset preview model is obtained.
[0026] Understandably, the target feature is a feature that the model can understand and has an impact on the strategy preview. The specific preset data screening methods are correlation analysis and statistical methods (ANOVA, chi-square test), where correlation analysis refers to the analysis of two or more correlated variable elements to measure the degree of correlation between two variable factors. ANOVA is a statistical formula used to compare the variance of the means (or averages) of different groups to determine whether there are significant differences between these means. The chi-square test is a hypothesis testing method, which is mainly used to determine whether there are significant differences between observed data and theoretical expectations, and then analyze whether two categorical variables are independent or related; the preset model structure can be the XGBoost model structure, where XGBoost (eXtreme Gradient Boosting) is an integrated learning algorithm and an improvement of the gradient boosting tree (GBDT). It builds a powerful integrated model by combining multiple weak learners (usually decision trees). In XGBoost, each weak learner is usually a decision tree (such as a CART tree), and these decision trees are integrated through an additive model, that is, each tree contributes to the final prediction result; Specifically, the successfully defined target variable Y-conversion rate is obtained in advance, the screened target features are used as the input of the preset model structure, the initial model parameters are set at the same time, and the model of the model is trained using a data set with a success rate label; the model parameters in the model structure are adjusted through cross-validation and grid search methods, and after determining that the model parameters meet the preset model parameters, a successfully trained preset preview model is obtained to ensure the generalization ability of the model; The target features screened out in this embodiment can better reflect the essential characteristics of the data, which helps the model learn more accurate prediction rules, thereby improving the prediction accuracy of the model and enabling the model to play a better role in practical applications; through reasonable feature selection and model parameter adjustment, the risk of overfitting of the model can be reduced, making the performance of the model on unknown data more stable and enhancing the generalization ability of the model.
[0027] Furthermore, obtaining the preview success rate corresponding to the data to be previewed of the rule strategy through different rule strategy parameters in the preset rule strategy preview model includes: Under different rule strategy parameters in the preset rule strategy preview model, after randomly sampling the data to be previewed for the rule strategy by a preset statistical simulation method, a plurality of simulation results on different rule strategies after random sampling are obtained; After comparing the simulation results of different rule strategies, the rehearsal success rate is obtained.
[0028] Understandably, rule strategy parameters refer to model-related parameters (such as eigenvectors that can be understood by the model) obtained after encoding and processing different rule strategies; the preset statistical simulation method can be the Monte Carlo method, which is a random number-based calculation method used to solve mathematical and physical problems. This method uses random sampling to perform numerical approximation or probabilistic statistics on the solution to the problem. The core idea is to use randomness to solve deterministic problems, or to use random experiments to estimate some values that are difficult to calculate directly; Specifically, for predefined rule strategies, the target features are converted into understandable model features according to the rule strategy encoding, and the target features are input into the preset rule strategy preview model; under different rule strategy parameters, based on the distribution of predicted conversion rate, random sampling is performed using a preset statistical simulation method, wherein each sampling represents an "experiment", and the average effect and possible fluctuation range of the rule strategy are estimated through the results of a large number of experiments; the conversion rate results of each simulation are recorded, and the average conversion rate and distribution of the conversion rate (such as standard deviation, confidence space, etc.) of each rule strategy are counted; by comparing the simulation results of different rule strategies, the average effect and risk of the rule strategy are evaluated; according to the simulation results, the rule strategy with the highest conversion rate, the most stable or the lowest risk is identified, and at the same time, the strategy parameters are dynamically adjusted to optimize the strategy combination to help improve the conversion rate or reduce the risk; This embodiment evaluates the rehearsal success rate of different rule strategies through random sampling and simulation experiments, improves the scientificity and effectiveness of model data processing, and automatically simulates the operational effects after the implementation and offline of the strategy, and can evaluate its expected impact on key indicators such as business growth and risk control.
[0029] Furthermore, before obtaining at least one total score corresponding to the data to be previewed for the rule strategy through the expected evaluation index in the preset index evaluation effectiveness model, the method further includes: Predefine and quantify the model evaluation index corresponding to the data to be previewed for the rule strategy; The indicator weights are allocated to the model evaluation indicators according to a preset hierarchical analysis method or a preset entropy weight method.
[0030] Understandably, the model evaluation index is the index corresponding to the rule strategy in the business process, including but not limited to conversion rate, expense rate, customer churn rate and other operational indicators that the business pays more attention to, such as conversion rate: the conversion rate prediction value directly output by the Monte Carlo simulation algorithm, expense rate: the total cost is calculated according to the strategy parameters (the services delivered, the promotion cost), and then the total insured amount is calculated according to the conversion rate prediction and the average insured amount, and the final expense rate = total cost / total insured amount, customer churn rate: set a threshold, and customers below this threshold are considered to have a low probability of insurance coverage and can be judged as potential lost customers, customer churn rate = potential lost customers / total number of customers; Specifically, the preset analytic hierarchy process assigns indicator weights to model evaluation indicators in the following way: decompose the problem of decision indicator weights into different levels, including the target level, the criterion level and the scheme level, where the target level is the final goal or ideal result of the decision problem, the criterion level is the various criteria or factors that need to be considered to achieve the goal, and the scheme level is the specific scheme or measure to achieve the goal; by comparing the importance of each factor in the same level to a certain criterion in the previous level, construct a judgment matrix, and the elements of the judgment matrix are assigned using the 1-9 scaling method to represent the relative importance of each factor; by solving the maximum eigenvalue of the judgment matrix and its corresponding eigenvector, the weight vector (i.e., indicator weight) of each factor for the criterion of the previous level is obtained, and the commonly used solution methods are the sum-product method and the square root method; The preset entropy weight method assigns indicator weights to the model evaluation indicators in the following manner: perform standardization, that is, convert the absolute value of the indicator into a relative value to solve the homogeneity problem of the values of various indicators of different qualities, wherein the commonly used data standardization methods include but are not limited to minimum-maximum standardization and Z-score standardization; calculate the information entropy of each indicator according to the information entropy calculation formula (Ej=-∑Pj*log2(Pj)), wherein Pj represents the proportion of each sample under the jth indicator, and Ej represents the information entropy of the jth indicator; Calculate the coefficient of difference, where the coefficient of difference is the inverse process of information entropy and is used to measure the degree of difference between indicators. The larger the coefficient of difference, the more discrete the data of the indicator, and the greater the impact on the comprehensive evaluation; calculate the weight of each indicator based on the coefficient of difference, where the size of the weight reflects the importance of each indicator in the comprehensive evaluation; It should be noted that this embodiment can pre-set the target layer according to the business objectives or use it to calculate the coefficient of difference, and set weights for each model evaluation indicator. For example, if the conversion rate is the most important goal, a higher weight can be set for the conversion rate. In general, by assigning indicator weights, a total score that meets the business objectives can be obtained, and the data results are more accurate.
[0031] Furthermore, the obtaining of the total score of the data to be previewed for the rule strategy includes: The total score of the data to be previewed for the rule strategy is obtained by weighted summation of the model evaluation index and the index weight; one model evaluation index corresponds to one index weight; Analyze the total score of each data to be previewed for rule strategies and the score of each model evaluation indicator, and identify and determine the improvement content and change trend of the data to be previewed for rule strategies.
[0032] Understandably, the score of each model evaluation indicator is multiplied by the indicator weight, and then the weighted scores are added to obtain the total score of the strategy rule; Specifically, for each rule strategy, the total score of the strategy is obtained by weighted summation of the indicator value and weight, where the total score = Σ(indicator value * indicator weight); the total score of each rule strategy and the score of each model evaluation indicator are analyzed to identify the best performing rule strategy and the content that needs to be improved. At the same time, according to the indicator analysis results, if the total score of a rule strategy remains unchanged or begins to show a downward trend, it can be considered that the effectiveness of the rule strategy has reached its upper limit, and it is determined that it needs to be optimized or replaced to ensure the continued effectiveness and competitiveness of the rule strategy; This embodiment calculates the total score by weighted summation and comprehensively considers the scores of each model evaluation index, so as to more comprehensively and accurately evaluate the performance of the data to be previewed for the rule strategy; by analyzing the total score and the scores of each model evaluation index, the improvement content and change trend of the data can be identified, thereby providing a scientific basis for the adjustment of the rule strategy; based on accurate evaluation results and scientific analysis, more reasonable and effective rule strategies can be formulated, thereby optimizing the decision-making process and improving the scientificity and rationality of the decision; by comparing and analyzing the total scores of different rule strategies, the strategy with better performance can be accurately located, thereby providing strong data support for the decision-making process; It should be noted that in one embodiment, a scoring threshold can be set. Once the score of a rule strategy is lower than this scoring threshold for a long period of time, it can be considered that the effectiveness of the strategy has reached its upper limit. At this time, optimization or replacement of the rule strategy should be considered to ensure the continued effectiveness and competitiveness of the rule strategy.
[0033] In summary, the above provides a model-based rule strategy preview method, which belongs to the field of intelligent decision-making technology, and obtains data to be previewed for rule strategy; after inputting the data to be previewed for rule strategy into a preset rule strategy preview model, the preview success rate and total score of the data to be previewed for rule strategy are obtained; the preset rule strategy preview model includes a preset preview model and a preset indicator evaluation effectiveness model; the processing method of the preset rule strategy preview model includes, after inputting the data to be previewed for rule strategy into the preset rule strategy preview model, obtaining the preview success rate corresponding to the data to be previewed for rule strategy and at least one expected evaluation indicator corresponding to the preview success rate through different rule strategy parameters in the preset rule strategy preview model; the processing method of the preset indicator evaluation effectiveness model includes obtaining at least one total score corresponding to the data to be previewed for rule strategy through the expected evaluation indicators in the preset indicator evaluation effectiveness model, and the expected evaluation indicators affect the model evaluation indicators in the preset indicator evaluation effectiveness model. This method introduces an intelligent preset rule strategy preview model to replace the original manual processing method. The model also specifically sets up a preset preview model for predicting the success rate and a preset indicator evaluation efficiency model for scoring. The model can quantitatively evaluate strategies and rules, which can improve the efficiency of decision-making or evaluation. Decision-making or evaluation does not require human participation, which can save time and cost. At the same time, the model's evaluation process can comprehensively evaluate the impact of multiple indicators in the operation process, provide a scientific basis for strategy optimization, effectively reduce operational risks, and ensure that enterprises can respond to market changes more accurately and efficiently, continuously improve their competitiveness, and inject new vitality into the healthy development of the industry. Through the established rule strategy preview model, Before the rule strategy is officially launched, its possible performance and effects can be predicted. This forward-looking ability can help decision makers make adjustments and optimizations during the strategy formulation stage to avoid potential operational risks. The rule strategy evaluation system converts the performance and effects of the rule strategy into quantifiable data by setting a series of evaluation indicators and standards, making the evaluation process more objective and fair, helping to more accurately measure the actual value of the strategy and provide data support for subsequent decision-making. Through the established rule strategy preview model, different environmental conditions can be simulated according to the input parameters, and the strategy performance in different scenarios can be tested. It can adapt to the ever-changing business needs and market environment, and can be expanded to other fields, with broad application potential.
[0034] It should be understood that the order of execution of the steps in the above embodiment does not necessarily mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present invention.
[0035] In one embodiment, a model-based rule strategy preview device is provided, and the model-based rule strategy preview device corresponds to the model-based rule strategy preview method in the above embodiment. Figure 3 As shown, the model-based rule strategy preview device includes an acquisition module 11 and an input module 12. Each functional module is described in detail as follows: An acquisition module is used to acquire data to be previewed for rule strategies; An input module is used to input the data to be previewed of the rule strategy into a preset rule strategy preview model, and then obtain the preview success rate and total score of the data to be previewed of the rule strategy; the preset rule strategy preview model includes a preset preview model and a preset indicator evaluation effectiveness model; the processing method of the preset rule strategy preview model includes, after the data to be previewed of the rule strategy is input into the preset rule strategy preview model, obtaining the preview success rate corresponding to the data to be previewed of the rule strategy and at least one expected evaluation indicator corresponding to the preview success rate through different rule strategy parameters in the preset rule strategy preview model; the processing method of the preset indicator evaluation effectiveness model includes obtaining at least one total score corresponding to the data to be previewed of the rule strategy through the expected evaluation indicators in the preset indicator evaluation effectiveness model, and the expected evaluation indicators affect the model evaluation indicators in the preset indicator evaluation effectiveness model.
[0036] Furthermore, the model-based rule strategy preview device includes: A collection module, used to collect historical data related to the rule strategy preview to be performed; The preprocessing module is used to perform data preprocessing on the historical data to obtain training data with outliers removed and / or value filling completed.
[0037] Furthermore, the model-based rule strategy preview device also includes: An encoding module, used for encoding the rule strategy parameters corresponding to the historical data to obtain encoded training data; The data conversion module, or / and, is used to perform data conversion on the preset type data corresponding to the historical data to obtain training data after data conversion.
[0038] Furthermore, the model-based rule strategy preview device also includes: A screening module, used to screen out target features from training data according to a preset data screening method; A confirmation module is used to input the target features into a preset model structure, and use training data with annotated success rate labels to train the preset model structure, and after confirming that the model parameters adjusted by the preset model parameter adjustment method meet the preset model parameters, the preset preview model is obtained.
[0039] Furthermore, the input module comprises: A random sampling submodule, used to obtain a plurality of simulation results on different rule strategies after random sampling of the data to be previewed for the rule strategy by a preset statistical simulation method under different rule strategy parameters in the preset rule strategy preview model; The comparison submodule is used to compare the simulation results of different rule strategies to obtain the preview success rate.
[0040] Furthermore, the model-based rule strategy preview device also includes: A quantification module, used to predefine and quantify the model evaluation index corresponding to the data to be previewed for the rule strategy; The allocation module is used to allocate indicator weights to the model evaluation indicators according to a preset hierarchical analysis method or a preset entropy weight method.
[0041] Furthermore, the input module comprises: A weighted summation submodule, used for obtaining the total score of the data to be previewed for the rule strategy by weighted summation according to the model evaluation index and the index weight; one model evaluation index corresponds to one index weight; A submodule is determined, which is used to analyze the total score of each data to be previewed for rule strategies and the score of each model evaluation index, and identify and determine the improvement content and change trend of the data to be previewed for rule strategies.
[0042] For the specific definition of the model-based rule strategy preview device, please refer to the definition of the model-based rule strategy preview method above, which will not be repeated here. Each module in the above-mentioned model-based rule strategy preview device can be implemented in whole or in part by software, hardware and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in the form of hardware, or can be stored in the memory of the computer device in the form of software, so that the processor can call and execute the operations corresponding to the above modules.
[0043] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 4As shown. The computer device includes a processor, a memory, a network interface and a database connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store data involved in the model-based rule strategy preview method. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a model-based rule strategy preview method is implemented.
[0044] In one embodiment, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps of the model-based rule strategy preview method in the above embodiment are implemented, such as Figure 2 Steps S10 to S20 shown include the following steps: Obtain data for previewing the rule strategy; After the data to be previewed for rule strategy are input into the preset rule strategy preview model, the preview success rate and total score of the data to be previewed for rule strategy are obtained; the preset rule strategy preview model includes a preset preview model and a preset indicator evaluation effectiveness model; the processing method of the preset rule strategy preview model includes, after the data to be previewed for rule strategy are input into the preset preview model, obtaining the preview success rate corresponding to the data to be previewed for rule strategy and at least one expected evaluation indicator corresponding to the preview success rate through different rule strategy parameters in the preset preview model; the processing method of the preset indicator evaluation effectiveness model includes obtaining at least one total score corresponding to the data to be previewed for rule strategy through the expected evaluation indicators in the preset indicator evaluation effectiveness model, and the expected evaluation indicators affect the model evaluation indicators in the preset indicator evaluation effectiveness model.
[0045] Alternatively, when the processor executes the computer program, the functions of each module / unit of the model-based rule strategy preview device in the above embodiment are realized, for example Figure 3 The functions of modules 11 to 12 are shown in Figure 1. To avoid repetition, they will not be described here.
[0046] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of the model-based rule strategy preview method in the above embodiment are implemented, such as Figure 2 Steps S10 to S20 shown include the following steps: Obtain data for previewing the rule strategy; After the data to be previewed for rule strategy are input into the preset rule strategy preview model, the preview success rate and total score of the data to be previewed for rule strategy are obtained; the preset rule strategy preview model includes a preset preview model and a preset indicator evaluation effectiveness model; the processing method of the preset rule strategy preview model includes, after the data to be previewed for rule strategy are input into the preset preview model, obtaining the preview success rate corresponding to the data to be previewed for rule strategy and at least one expected evaluation indicator corresponding to the preview success rate through different rule strategy parameters in the preset preview model; the processing method of the preset indicator evaluation effectiveness model includes obtaining at least one total score corresponding to the data to be previewed for rule strategy through the expected evaluation indicators in the preset indicator evaluation effectiveness model, and the expected evaluation indicators affect the model evaluation indicators in the preset indicator evaluation effectiveness model.
[0047] Alternatively, when the computer program is executed by the processor, the functions of each module / unit of the model-based rule strategy preview device in the above embodiment are realized, for example Figure 3 The functions of modules 11 to 12 are shown in Figure 1. To avoid repetition, they will not be described here.
[0048] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0049] Those skilled in the art can clearly understand that for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional units and modules as needed, that is, the internal structure of the device can be divided into different functional units or modules to complete all or part of the functions described above.
[0050] The embodiments described above are only used to illustrate the technical solutions of the present invention, rather than to limit the same. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention, and should all be included in the protection scope of the present invention.
Claims
1. A model-based rule strategy preview method, characterized in that: include: Obtain data for previewing the rule strategy; After the data to be previewed for rule strategy are input into the preset rule strategy preview model, the preview success rate and total score of the data to be previewed for rule strategy are obtained; the preset rule strategy preview model includes a preset preview model and a preset indicator evaluation effectiveness model; the processing method of the preset rule strategy preview model includes, after the data to be previewed for rule strategy are input into the preset preview model, obtaining the preview success rate corresponding to the data to be previewed for rule strategy and at least one expected evaluation indicator corresponding to the preview success rate through different rule strategy parameters in the preset preview model; the processing method of the preset indicator evaluation effectiveness model includes obtaining at least one total score corresponding to the data to be previewed for rule strategy through the expected evaluation indicators in the preset indicator evaluation effectiveness model, and the expected evaluation indicators affect the model evaluation indicators in the preset indicator evaluation effectiveness model.
2. The model-based rule strategy preview method according to claim 1, characterized in that: After inputting the data to be previewed for the rule strategy into a preset rule strategy preview model, before obtaining the preview success rate and total score of the data to be previewed for the rule strategy, the method further includes: Collect historical data related to the rule strategy preview to be conducted; The historical data is preprocessed to obtain training data with outliers removed and / or value filling completed.
3. The model-based rule strategy preview method according to claim 2, characterized in that: Before performing data preprocessing on the historical data to obtain training data with outliers removed and / or value filling completed, the method further includes: Encoding the rule strategy parameters corresponding to the historical data to obtain encoded training data; Or / and, performing data conversion on preset type data corresponding to the historical data to obtain training data after data conversion.
4. The model-based rule strategy preview method according to any one of claims 1 to 3, characterized in that: Before obtaining the preview success rate corresponding to the data to be previewed for the rule strategy and at least one expected evaluation index corresponding to the preview success rate through different rule strategy parameters in the preset rule strategy preview model, the method further includes: Filter out target features from the training data according to a preset data filtering method; The target features are input into a preset model structure, and the preset model structure is trained using training data with annotated success rate labels. After confirming that the model parameters adjusted by a preset model parameter adjustment method meet the preset model parameters, the preset preview model is obtained.
5. The model-based rule strategy preview method according to claim 1, characterized in that: The obtaining of the preview success rate corresponding to the data to be previewed of the rule strategy through different rule strategy parameters in the preset rule strategy preview model includes: Under different rule strategy parameters in the preset rule strategy preview model, after randomly sampling the data to be previewed for the rule strategy by a preset statistical simulation method, a plurality of simulation results on different rule strategies after random sampling are obtained; After comparing the simulation results of different rule strategies, the rehearsal success rate is obtained.
6. The model-based rule strategy preview method according to any one of claims 1 to 3, characterized in that: Before obtaining at least one total score corresponding to the data to be previewed of the rule strategy through the expected evaluation index in the preset index evaluation effectiveness model, the method further includes: Predefine and quantify the model evaluation index corresponding to the data to be previewed for the rule strategy; The indicator weights are allocated to the model evaluation indicators according to a preset hierarchical analysis method or a preset entropy weight method.
7. The model-based rule strategy preview method according to claim 1, characterized in that: The obtaining of the total score of the data to be previewed for the rule strategy includes: The total score of the data to be previewed for the rule strategy is obtained by weighted summation of the model evaluation index and the index weight; one model evaluation index corresponds to one index weight; Analyze the total score of each data to be previewed for rule strategies and the score of each model evaluation indicator, and identify and determine the improvement content and change trend of the data to be previewed for rule strategies.
8. A model-based rule strategy preview device, characterized in that: include: An acquisition module is used to acquire data to be previewed for rule strategies; An input module is used to input the data to be previewed of the rule strategy into a preset rule strategy preview model, and then obtain the preview success rate and total score of the data to be previewed of the rule strategy; the preset rule strategy preview model includes a preset preview model and a preset indicator evaluation effectiveness model; the processing method of the preset rule strategy preview model includes, after the data to be previewed of the rule strategy is input into the preset rule strategy preview model, obtaining the preview success rate corresponding to the data to be previewed of the rule strategy and at least one expected evaluation indicator corresponding to the preview success rate through different rule strategy parameters in the preset rule strategy preview model; the processing method of the preset indicator evaluation effectiveness model includes obtaining at least one total score corresponding to the data to be previewed of the rule strategy through the expected evaluation indicators in the preset indicator evaluation effectiveness model, and the expected evaluation indicators affect the model evaluation indicators in the preset indicator evaluation effectiveness model.
9. A computer device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the model-based rule strategy preview method according to any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the model-based rule strategy preview method according to any one of claims 1 to 7 is implemented.