Reservoir immigrant support fund performance evaluation system and method thereof

Abstract

The invention discloses a reservoir immigrant support fund performance evaluation system and method, and particularly relates to the technical field of social governance and support fund evaluation. On the basis of household registration information, community visit data and a big data analysis technology, a classification model of a target support group is constructed, and immigrant groups influenced by reservoir construction are accurately identified; acquiring income data, employment state data and educational resource data of the target group, calculating key performance indicators of the target group by adopting algorithms such as income growth rate calculation, employment rate statistics and educational resource coverage rate calculation, and constructing a support fund performance evaluation model according to an evaluation result; according to an evaluation result, dynamically adjusting a distribution strategy of supporting funds, and preferentially distributing the funds to a group of which the performance does not reach the standard; a dynamic monitoring mechanism is constructed, and the problems in the support policy are identified and adjusted in time by continuously tracking the life quality and policy implementation effect of the target group, so that the configuration of support resources is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of social governance and support fund evaluation, and in particular to a reservoir migrant support fund performance evaluation system and method. Background Art

[0002] With the advancement of reservoir construction projects, many migrant groups have been affected to varying degrees due to land expropriation, environmental changes and other reasons, and their quality of life and livelihood conditions are facing severe challenges. In order to guarantee the basic living needs of these migrant groups and promote their long-term development, the government and relevant departments have introduced support funds and policies. However, in the traditional allocation of support funds and policy implementation, there are often problems such as uneven resource allocation and unclear support effects. Due to the lack of accurate evaluation and adjustment mechanisms, many affected migrant groups have not received timely and effective help, which has affected the actual effect of support policies and social justice. Therefore, how to accurately identify target groups, scientifically evaluate the performance of support funds, and timely adjust resource allocation strategies have become key issues that need to be solved.

[0003] What the existing technology lacks: Combining digital processing technology, big data analysis and intelligent decision-making models, through precise classification and real-time monitoring, to ensure the rational allocation of support resources. Specifically, the present invention first uses household registration information, community visit data and external data sources to accurately identify the target groups affected by reservoir construction through a decision tree algorithm, and constructs a support fund performance evaluation model based on indicators such as income growth rate, employment rate and education resource coverage. Through this model, it is possible to comprehensively evaluate whether the support funds meet the preset standards, and on this basis, dynamically adjust the allocation strategy of support funds. The system also continuously tracks the effectiveness of policy implementation through a continuous dynamic monitoring mechanism, and uses feedback analysis to timely adjust support measures, thereby achieving refined management and optimization of support policies and improving the efficiency and effectiveness of policy implementation. Summary of the invention

[0004] The purpose of the present invention is to provide a reservoir migrant support fund performance evaluation system and method to solve the above-mentioned background problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] The performance evaluation method of reservoir resettlement support funds includes the following steps:

[0007] S1: Based on the digital processing of household registration information, community visit data collection and big data analysis technology, a classification model for the target support group is constructed, and the immigrant group affected by the reservoir construction is identified according to the classification model, recorded as the target group, and the classification data of the target group is generated;

[0008] S2: Obtain the income data, employment status data and education resource data of the target group, calculate the income growth rate, employment rate and education resource coverage rate of the target group using the income growth rate calculation algorithm, employment rate statistical algorithm and education resource coverage rate calculation algorithm, and build a support fund performance evaluation model based on the calculation results, and evaluate whether the support fund performance indicators meet the preset standards according to the evaluation model;

[0009] S3: Dynamically adjust the allocation strategy of support funds according to the evaluation results of the support fund performance evaluation model, and the adjustment includes:

[0010] Set funding allocation priority parameters, determine the priority allocation direction of support resources, and allocate support funds to target groups whose performance indicators do not meet the preset standards;

[0011] Implement the planning and allocation of skills training resources, the optimal configuration of educational facilities and the adjustment of living allowance programs;

[0012] S4: Establish a dynamic monitoring mechanism to track the quality of life of the target groups and the effectiveness of the implementation of support policies, and use feedback analysis methods to identify problems that arise during policy implementation and generate adjustment strategies.

[0013] As a further solution of the present invention: the identification of the immigrant groups affected by the reservoir construction according to the classification model, recorded as the target group, specifically includes:

[0014] Collect big data from household registration information, community visit data, and external data sources;

[0015] The collected household registration information, community visit data and big data from external data sources are converted into a computable table form to generate a data set D = {X i ,Y};

[0016] In the formula, X i represents the eigenvector of an individual, including variables of household income, land area, and education level, and i represents the number of eigenvectors;

[0017] Y indicates whether the individual meets the support conditions;

[0018] Ensure data quality by handling missing values, outliers, and duplicate data, including:

[0019] The missing values ​​were filled using the mean interpolation method, and outliers were removed using the box plot method or the standard deviation method;

[0020] Based on data set D, a classification model for the target support group is constructed using a classification algorithm;

[0021] Wherein, the classification model is a decision tree model;

[0022] When selecting classification features, calculate the information gain IG(X), the calculation expression is: IG(X) = H(Y)-H(Y|X);

[0023] Among them, H(Y) represents the entropy of the target group label, and the calculation expression is: H(Y) = - H(Y|X) represents the classification entropy under given conditions of feature X, and the calculation expression is:

[0024] In the formula, P(y i ) represents the target classification label y i The probability of k represents the total number of classification labels, P(x i ) indicates that the feature X takes the value x i The probability of H(Y|x i ) indicates that the feature X takes the value x i The classification entropy of

[0025] Select the feature X with the maximum information gain from the feature set * ;

[0026] Use recursive method to build decision tree, make information gain less than preset threshold, classification error lower than preset threshold, and generate classification rules;

[0027] Use the classification model to predict individual data, obtain classification results, and determine the target group based on the classification results.

[0028] As a further solution of the present invention: the income growth rate calculation algorithm, employment rate statistical algorithm and education resource coverage rate calculation algorithm are used to calculate the income growth rate, employment rate and education resource coverage rate of the target group, and a support fund performance evaluation model is constructed based on the calculation results to calculate the comprehensive performance score, which specifically includes:

[0029] Realize real-time monitoring of target group data through the big data platform, and obtain income, employment and education resource data from various data sources in real time;

[0030] Standardize the target group's income data, employment status data, and educational resource data;

[0031] Divide the standardized data into training set and test set;

[0032] Constructing an indicator prediction model, wherein the indicator prediction model is a random forest model;

[0033] Among them, the training process of the random forest model is:

[0034] The random forest regression model consists of multiple decision trees, and the model output calculation expression is:

[0035]

[0036] In the formula, represents the prediction result of the random forest model, Z represents the input feature vector, including individual income, employment data and educational resources, M represents the maximum number of decision trees in the random forest, and f j (Z) represents the prediction result of the jth decision tree, where j represents the number of decision trees in the random forest;

[0037] Use the loss function to optimize the random forest model;

[0038] Using mean square error as the loss function, the calculation expression is:

[0039] In the formula, i represents the number of eigenvectors, y i The real values ​​include: income growth rate, employment rate, and educational resource coverage rate. Indicates that the model prediction values ​​include: income growth rate, employment rate, and educational resource coverage rate. a represents the number of training samples, and N represents the maximum number of training samples.

[0040] According to the trained random forest model, the income growth rate, employment rate, and educational resource coverage rate are predicted respectively. The predicted income growth rate, employment rate, and educational resource coverage rate are normalized to calculate the comprehensive performance score.

[0041] As a further solution of the present invention: the step of evaluating whether the performance indicators of the support funds have reached the preset standards according to the evaluation model specifically includes:

[0042] Determine whether the comprehensive performance score is greater than or equal to the preset threshold. If so, the support fund performance indicator reaches the preset standard. If not, the support fund performance indicator does not reach the preset standard.

[0043] As a further solution of the present invention: the process of obtaining the fund allocation priority parameter is as follows:

[0044] The comprehensive performance score of each target group is calculated by comparing it with the predicted income growth rate, employment rate, and education resource coverage rate. The calculated results are summed up to obtain the priority score of each target group.

[0045] As a further solution of the present invention: the determination of the priority allocation direction of the support resources specifically includes:

[0046] Compare the priority score with the first threshold. If the priority score is greater than or equal to the first threshold, the preset standard deviation of the support resources and support fund performance indicators of the corresponding target group is large, which is recorded as high priority. If the priority score is less than the first threshold and greater than the second threshold, the preset standard deviation of the support resources and support fund performance indicators of the corresponding target group is average, which is recorded as medium priority. If the priority score is less than the second threshold, the preset standard deviation of the support resources and support fund performance indicators of the corresponding target group is small, which is recorded as low priority.

[0047] The first threshold is greater than the second threshold.

[0048] As a further solution of the present invention: the dynamic monitoring mechanism is constructed, by tracking the quality of life of the target group and the implementation effect of the support policy, using the feedback analysis method to identify the problems that arise in the policy implementation process and generate adjustment strategies, specifically including:

[0049] By collecting income data, employment data and educational resource data of the target group according to the monitoring cycle, and standardizing the collected data, the monitoring indicators and deviations of the group are calculated;

[0050] Judging the quality of life and the effectiveness of policy implementation, including:

[0051] Income bias, employment bias, and education coverage bias are calculated separately;

[0052] Obtain the sum of weighted deviations of all target groups to obtain an overall execution score, and compare the overall execution score with a preset threshold. If the overall execution score is greater than or equal to the preset threshold, the overall execution status is good; if the overall execution score is less than the preset threshold, the overall execution status has problems;

[0053] Based on the problems in the overall implementation status, feedback analysis algorithms are used to identify problem areas in policy implementation, and regression models are used to analyze the relationship between various policy indicators and deviations to determine the specific factors that affect policy implementation;

[0054] By analyzing the deviation values, adjustment strategies are generated to optimize the allocation of supporting funds and resources.

[0055] The performance evaluation system of reservoir resettlement support funds includes:

[0056] A data collection module, which is used to obtain income data, employment status data, educational resource data of the target group, collect household registration information, community visit data and big data from external data sources;

[0057] A target group determination module, which constructs a classification model for the target support group based on digital processing of household registration information, community visit data collection and big data analysis technology, and identifies the immigrant group affected by the reservoir construction according to the classification model and records it as the target group;

[0058] A support fund performance indicator judgment module, which uses an income growth rate calculation algorithm, an employment rate statistical algorithm, and an education resource coverage rate calculation algorithm to calculate the income growth rate, employment rate, and education resource coverage rate of the target group, and builds a support fund performance evaluation model based on the calculation results, and evaluates whether the support fund performance indicators meet the preset standards according to the evaluation model;

[0059] A dynamic adjustment module, which dynamically adjusts the allocation strategy of the support funds according to the evaluation results of the support fund performance evaluation model;

[0060] The dynamic monitoring module builds a dynamic monitoring mechanism, tracks the quality of life of the target group and the implementation effect of the support policy, and uses feedback analysis methods to identify problems that arise during policy implementation and generate adjustment strategies.

[0061] Beneficial effects of the present invention:

[0062] (1) The present invention constructs a classification model for the target support group based on digital processing of household registration information, community visit data collection and big data analysis technology to identify the immigrant groups affected by reservoir construction. Through data cleaning and processing, the decision tree classification algorithm is used to accurately divide the target group, ensuring that each immigrant family or individual can be accurately identified as to whether they meet the support conditions. This enables the support funds to be more efficiently allocated to the groups in need of support. The results of accurate identification not only improve the efficiency of the use of support funds, but also ensure that the support policy is highly targeted and the resource allocation is more reasonable. The classification model can continuously adjust the division criteria of the target group according to dynamically changing external data, ensuring the adaptability and flexibility of the support policy. Through this scientific and reasonable support group identification method, the accuracy and effect of policy implementation are improved, and the real needs of the immigrant groups affected by reservoir construction can be met to the greatest extent.

[0063] (2) The dynamic monitoring mechanism in the present invention can track the quality of life of the target group and the implementation effect of the support policy in real time, providing strong support for the continuous optimization of the policy. Through the real-time data collection and analysis of the big data platform, the income growth, employment status and educational resource coverage of the target group can be regularly evaluated, and the deviations and problems in the policy implementation process can be discovered in time. For example, by calculating the income deviation, employment deviation and education coverage deviation, the gap between the actual benefits of the support funds and resources and the expected goals can be clearly reflected. When the monitoring results show that a certain policy indicator does not meet the preset standard, the system can identify the problem areas in the implementation through the feedback analysis algorithm and generate corresponding adjustment strategies. This mechanism not only enhances the operability and flexibility of policy implementation, but also ensures that the support funds and resources can be dynamically adjusted according to the specific needs of the target group, making the support effect more accurate and sustainable. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The present invention will be further described below in conjunction with the accompanying drawings.

[0065] Figure 1 It is a flowchart of the specific steps of the performance evaluation method of reservoir immigrant support funds of the present invention;

[0066] Figure 2 It is a flow chart of the reservoir immigrant support fund performance evaluation system of the present invention. DETAILED DESCRIPTION

[0067] 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 only 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.

[0068] See also Figure 1 As shown, the present invention is a performance evaluation method for reservoir migrant support funds, comprising the following steps:

[0069] S1: Based on the digital processing of household registration information, community visit data collection and big data analysis technology, a classification model for the target support group is constructed, and the immigrant group affected by the reservoir construction is identified according to the classification model, recorded as the target group, and the classification data of the target group is generated;

[0070] S2: Obtain the income data, employment status data and education resource data of the target group, calculate the income growth rate, employment rate and education resource coverage rate of the target group using the income growth rate calculation algorithm, employment rate statistical algorithm and education resource coverage rate calculation algorithm, and build a support fund performance evaluation model based on the calculation results, and evaluate whether the support fund performance indicators meet the preset standards according to the evaluation model;

[0071] S3: Dynamically adjust the allocation strategy of support funds according to the evaluation results of the support fund performance evaluation model, and the adjustment includes:

[0072] Set funding allocation priority parameters, determine the priority allocation direction of support resources, and allocate support funds to target groups whose performance indicators do not meet the preset standards;

[0073] Implement reasonable planning and allocation of skills training resources, optimize the configuration of educational facilities and adjust living allowance programs to improve the coverage of support resources;

[0074] S4: Establish a dynamic monitoring mechanism to continuously track the quality of life of the target groups and the implementation effects of support policies, and use feedback analysis methods to identify problems that arise during policy implementation and generate adjustment strategies.

[0075] In S1, based on the digital processing of household registration information, community visit data collection and big data analysis technology, a classification model for the target support group is constructed. According to the classification model, the immigrant group affected by the reservoir construction is identified and recorded as the target group, and the classification data of the target group is generated, including:

[0076] Collect big data from household registration information, community visit data, and external data sources;

[0077] The collected household registration information, community visit data and big data from external data sources are converted into a computable table form to generate a data set D = {X i , Y};

[0078] Where, X i represents the eigenvector of an individual, including variables of household income, land area, and education level, and i represents the number of eigenvectors;

[0079] Y indicates whether the individual meets the support conditions (Y = 0 for non-compliance, Y = 1 for compliance);

[0080] Ensure data quality by handling missing values, outliers, and duplicate data, including:

[0081] The missing values ​​were filled using the mean interpolation method, and outliers were removed using the box plot method or the standard deviation method;

[0082] Based on data set D, a classification model for the target support group is constructed using a classification algorithm;

[0083] Wherein, the classification model is a decision tree model;

[0084] When selecting classification features, calculate the information gain IG(X), the calculation expression is: IG(X) = H(Y)-H(Y|X);

[0085] Among them, H(Y) represents the entropy of the target group label, and the calculation expression is: H(Y|X) represents the classification entropy under given conditions of feature X, and the calculation expression is:

[0086] In the formula, P(y i ) represents the target classification label y i The probability of k represents the total number of classification labels, P(x i ) indicates that the feature X takes the value x i The probability of H(Y|x i ) indicates that the feature X takes the value x i The classification entropy of

[0087] Select the feature X with the maximum information gain from the feature set * ;

[0088] Use recursive method to build decision tree until the information gain is less than the preset threshold, the classification error is lower than the preset threshold, and generate classification rules;

[0089] Use the classification model to predict individual data, obtain classification results, and determine the target group based on the classification results;

[0090] The classification results are stored as target group classification data;

[0091] It should be noted that the process of establishing the classification model recursively generates a decision tree and selects the feature with the largest information gain for data division until the termination condition is met; the generated classification rules ensure that the model can accurately identify the target group and support the precise implementation of subsequent support policies.

[0092] In S2, the income data, employment status data and education resource data of the target group are obtained, and the income growth rate calculation algorithm, employment rate statistical algorithm and education resource coverage rate calculation algorithm are used to calculate the income growth rate, employment rate and education resource coverage rate of the target group, and a support fund performance evaluation model is constructed based on the calculation results. According to the evaluation model, whether the support fund performance indicators meet the preset standards is evaluated, including:

[0093] Realize real-time monitoring of target group data through the big data platform, obtain income, employment and education resource data from various data sources in real time; process these data in real time and push the analysis results to the decision-making system in real time;

[0094] Standardize the target group's income data, employment status data, and educational resource data;

[0095] Divide the standardized data into training set and test set (ratio is 80% and 20%);

[0096] Constructing an indicator prediction model, wherein the indicator prediction model is a random forest model;

[0097] Among them, the training process of the random forest model is:

[0098] The random forest regression model consists of multiple decision trees, and the model output calculation expression is:

[0099]

[0100] In the formula, represents the prediction result of the random forest model, Z represents the input feature vector, including individual income, employment data and educational resources, M represents the maximum number of decision trees in the random forest, and f j (Z) represents the prediction result of the jth decision tree, where j represents the number of decision trees in the random forest;

[0101] Use the loss function to optimize the random forest model;

[0102] Using mean square error as the loss function, the calculation expression is:

[0103] In the formula, y i represents the real value (including income growth rate, employment rate, and educational resource coverage rate), represents the model prediction value, a represents the number of training samples, and N represents the maximum number of training samples;

[0104] According to the trained random forest model, the income growth rate, employment rate, and educational resource coverage rate are predicted respectively. The predicted income growth rate, employment rate, and educational resource coverage rate are normalized and the comprehensive performance score is calculated. The calculation expression is:

[0105] Where Pb represents the comprehensive performance score of the b-th predicted value, b represents the number of predicted values, and G Rb represents the revenue growth rate of the b-th forecast value, E Rb represents the employment rate of the b-th predicted value, C Ebrepresents the educational resource coverage rate of the b-th predicted value, w 1 、w 2 and w 3 is the preset weight factor;

[0106] comparing the composite performance score to a preset threshold;

[0107] If the comprehensive performance score is greater than or equal to the preset threshold, it means that the performance index of the support funds has reached the preset standard;

[0108] If the comprehensive performance score is less than the preset threshold, it means that the performance indicators of the support funds have not reached the preset standards;

[0109] It should be noted that the comprehensive performance score reflects whether the performance indicators of the support funds have reached the preset standards, and the larger the value of the comprehensive performance score, the more likely the corresponding support funds performance indicators have reached the preset standards.

[0110] In S3, according to the evaluation results of the support fund performance evaluation model, the allocation strategy of the support fund is dynamically adjusted, and the adjustment includes:

[0111] Set funding allocation priority parameters, determine the priority allocation direction of support resources, and allocate support funds to target groups whose performance indicators do not meet the preset standards;

[0112] The acquisition process of the fund allocation priority parameter is as follows:

[0113] Calculate the ratio of the comprehensive performance score of each target group to the predicted income growth rate, employment rate, and education resource coverage rate, and sum the calculated results to obtain the priority score of each target group;

[0114] Compare the priority score with the first threshold. If the priority score is greater than or equal to the first threshold, it means that the preset standard deviation of the support resources and support fund performance indicators of the corresponding target group is large, and it is recorded as high priority. If the priority score is less than the first threshold and greater than the second threshold, it means that the preset standard deviation of the support resources and support fund performance indicators of the corresponding target group is average, and it is recorded as medium priority. If the priority score is less than the second threshold, it means that the preset standard deviation of the support resources and support fund performance indicators of the target group is small, and it is recorded as low priority.

[0115] wherein the first threshold is greater than the second threshold;

[0116] Resources are allocated to target groups based on allocation priorities, including skills training allocation, education resource allocation and living allowance allocation.

[0117] In S4, a dynamic monitoring mechanism is established to continuously track the quality of life of the target group and the implementation effect of the support policy, and use feedback analysis methods to identify problems in the policy implementation process and generate adjustment strategies, including:

[0118] By collecting income data, employment data and educational resource data of the target group according to the monitoring cycle (such as every quarter), and standardizing the collected data, calculating the monitoring indicators and deviations of the group, and evaluating the quality of life and policy implementation effects, specifically including:

[0119] Income bias, employment bias, and education coverage bias are calculated separately;

[0120] Obtain the sum of weighted deviations of all target groups to obtain the overall execution score, and compare the overall execution score with the preset threshold. If the overall execution score is greater than or equal to the preset threshold, it means that the overall execution status is good; if the overall execution score is less than the preset threshold, it means that there are problems with the overall execution status;

[0121] Based on the problems in the overall implementation status, feedback analysis algorithms are used to identify problem areas in policy implementation, and regression models are used to analyze the relationship between various policy indicators and deviations to determine the specific factors that affect policy implementation;

[0122] By analyzing the deviation values, we can generate adjustment strategies and optimize the allocation of supporting funds and resources;

[0123] Exemplary; if the income growth rate does not meet the target, increase the funding allocation for skills training and employment support policies;

[0124] If the educational resource coverage rate is lower than the target, adjust the allocation strategy of educational infrastructure;

[0125] Apply the generated adjustment strategies to actual operations and adjust the allocation of support funds and resources;

[0126] It should be noted that the dynamic monitoring mechanism can not only track the quality of life of the target group and the effect of the support policy in real time, but also timely identify problems and adjust strategies through feedback analysis. Ultimately, support funds and resources can be allocated more accurately, improving the implementation effect of the policy.

[0127] See also Figure 2 As shown in the figure, the performance evaluation system of reservoir resettlement support funds includes:

[0128] A data collection module, which is used to obtain income data, employment status data, educational resource data of the target group, collect household registration information, community visit data and big data from external data sources;

[0129] A target group determination module, which constructs a classification model for the target support group based on digital processing of household registration information, community visit data collection and big data analysis technology, and identifies the immigrant group affected by the reservoir construction according to the classification model and records it as the target group;

[0130] A support fund performance indicator judgment module, which uses an income growth rate calculation algorithm, an employment rate statistical algorithm, and an education resource coverage rate calculation algorithm to calculate the income growth rate, employment rate, and education resource coverage rate of the target group, and builds a support fund performance evaluation model based on the calculation results, and evaluates whether the support fund performance indicators meet the preset standards according to the evaluation model;

[0131] A dynamic adjustment module, which dynamically adjusts the allocation strategy of the support funds according to the evaluation results of the support fund performance evaluation model;

[0132] The dynamic monitoring module builds a dynamic monitoring mechanism, tracks the quality of life of the target group and the implementation effect of the support policy, and uses feedback analysis methods to identify problems that arise during policy implementation and generate adjustment strategies.

[0133] The working principle of the present invention is as follows: based on the digital processing of household registration information, community visit data collection and big data analysis technology, a classification model of the target support group is constructed to identify the immigrant group affected by the reservoir construction and generate the classification data of the target group; then, the income data, employment status data and education resource data of the target group are obtained, and various indicators are calculated by using the income growth rate calculation algorithm, the employment rate statistical algorithm and the education resource coverage calculation algorithm, and a support fund performance evaluation model is constructed based on the calculation results, and it is evaluated whether the fund performance indicators meet the preset standards; if the standards are not met, the fund allocation strategy is dynamically adjusted, the resource allocation direction is determined by priority scoring, and the reasonable allocation of skill training resources, the optimal configuration of educational facilities and the adjustment of living allowance plans are implemented; finally, a dynamic monitoring mechanism is constructed, the target group data is regularly collected, the monitoring indicator deviation is calculated and the policy implementation problems are identified by feedback analysis methods, and the adjustment strategy is dynamically generated to optimize resource allocation; the present invention ensures the accurate investment of support funds and improves the policy implementation effect through the coordinated operation of classification, evaluation, adjustment and monitoring.

[0134] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0135] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0136] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0137] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not 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 embodiments of the present application.

[0138] The above is a detailed description of an embodiment of the present invention, but the content is only a preferred embodiment of the present invention and cannot be considered to limit the scope of implementation of the present invention. All equivalent changes and improvements made within the scope of the present invention should still fall within the scope of the patent coverage of the present invention.

Claims

1. The performance evaluation method of reservoir resettlement support funds is characterized by: The following steps are involved: S1: Based on the digital processing of household registration information, community visit data collection and big data analysis technology, a classification model for the target support group is constructed, and the immigrant group affected by the reservoir construction is identified according to the classification model, recorded as the target group, and the classification data of the target group is generated; S2: Obtain the income data, employment status data and education resource data of the target group, calculate the income growth rate, employment rate and education resource coverage rate of the target group using the income growth rate calculation algorithm, employment rate statistical algorithm and education resource coverage rate calculation algorithm, and build a support fund performance evaluation model based on the calculation results, and evaluate whether the support fund performance indicators meet the preset standards according to the evaluation model; S3: Dynamically adjust the allocation strategy of support funds according to the evaluation results of the support fund performance evaluation model, and the adjustment includes: Set funding allocation priority parameters, determine the priority allocation direction of support resources, and allocate support funds to target groups whose performance indicators do not meet the preset standards; Implement the planning and allocation of skills training resources, the optimal configuration of educational facilities and the adjustment of living allowance programs; S4: Establish a dynamic monitoring mechanism to track the quality of life of the target groups and the effectiveness of the implementation of support policies, and use feedback analysis methods to identify problems that arise during policy implementation and generate adjustment strategies.

2. The performance evaluation method for reservoir migrant support funds according to claim 1 is characterized in that: The immigrant groups affected by the reservoir construction are identified according to the classification model and recorded as target groups, including: Collect household registration information, community visit data and big data from external data sources; convert the collected household registration information, community visit data and big data from external data sources into a computable table form to generate a data set D = {X i ,Y}; where X i The feature vector representing an individual includes variables such as household income, land area, and education level, and i represents the number of feature vectors; Y represents whether an individual meets the support conditions; data quality is ensured by processing missing values, outliers, and duplicate data, including: using the mean interpolation method to fill missing values, and using the box plot method or the standard deviation method to remove outliers; based on the data set D, a classification algorithm is used to build a classification model for the target support group; wherein the classification model is a decision tree model; when selecting classification features, the information gain IG(X) is calculated, and the calculation expression is: IG(X) = H(Y) - H(Y|X); wherein H(Y) represents the entropy of the target group label, and the calculation expression is: H ( Y|X) represents the classification entropy under given conditions of feature X, and the calculation expression is: In the formula, P(y i ) represents the target classification label y i The probability of k represents the total number of classification labels, P(x i ) indicates that the feature X takes the value x i The probability of H(Y|x i ) indicates that the feature X takes the value x i The classification entropy when ; Select the feature X with the maximum information gain from the feature set * ; Use recursive method to establish decision tree, make information gain less than preset threshold, classification error lower than preset threshold, and generate classification rules; use classification model to predict individual data, obtain classification results, and determine target group according to classification results.

3. The performance evaluation method for reservoir migrant support funds according to claim 1 is characterized in that: The income growth rate calculation algorithm, employment rate statistical algorithm and education resource coverage rate calculation algorithm are used to calculate the income growth rate, employment rate and education resource coverage rate of the target group, and a support fund performance evaluation model is constructed based on the calculation results to calculate the comprehensive performance score, which specifically includes: The target group data is monitored in real time through the big data platform, and the income, employment and education resource data from various data sources are obtained in real time; the target group's income data, employment status data and education resource data are standardized; the standardized data are divided into a training set and a test set; an indicator prediction model is constructed, and the indicator prediction model is a random forest model; wherein the training process of the random forest model is: the random forest regression model is composed of multiple decision trees, and the model output calculation expression is: In the formula, represents the prediction result of the random forest model, Z represents the input feature vector, including individual income, employment data and educational resources, M represents the maximum number of decision trees in the random forest, and f j (Z) represents the prediction result of the jth decision tree, j represents the number of decision trees in the random forest; the random forest model is optimized using the loss function; the mean square error is used as the loss function, and the calculation expression is: In the formula, i represents the number of eigenvectors, y i The real values ​​include: income growth rate, employment rate, and educational resource coverage rate. Indicates that the model prediction values ​​include: income growth rate, employment rate, and educational resource coverage rate. a represents the number of training samples, and N represents the maximum number of training samples. According to the trained random forest model, the income growth rate, employment rate, and educational resource coverage rate are predicted respectively. The predicted income growth rate, employment rate, and educational resource coverage rate are normalized to calculate the comprehensive performance score.

4. The performance evaluation method for reservoir migrant support funds according to claim 1 is characterized in that: The evaluation model is used to evaluate whether the performance indicators of the support funds meet the preset standards, including: Determine whether the comprehensive performance score is greater than or equal to the preset threshold. If so, the support fund performance indicator reaches the preset standard. If not, the support fund performance indicator does not reach the preset standard.

5. The performance evaluation method for reservoir migrant support funds according to claim 1 is characterized in that: The process of obtaining the fund allocation priority parameter is as follows: The comprehensive performance score of each target group is calculated by comparing it with the predicted income growth rate, employment rate, and education resource coverage rate. The calculated results are summed up to obtain the priority score of each target group.

6. The performance evaluation method for reservoir migrant support funds according to claim 1 is characterized in that: Determining the priority allocation direction of support resources specifically includes: Compare the priority score with the first threshold. If the priority score is greater than or equal to the first threshold, the preset standard deviation of the support resources and support fund performance indicators of the corresponding target group is large, which is recorded as high priority. If the priority score is less than the first threshold and greater than the second threshold, the preset standard deviation of the support resources and support fund performance indicators of the corresponding target group is average, which is recorded as medium priority. If the priority score is less than the second threshold, the preset standard deviation of the support resources and support fund performance indicators of the corresponding target group is small, which is recorded as low priority. The first threshold is greater than the second threshold.

7. The performance evaluation method for reservoir migrant support funds according to claim 1 is characterized in that: The dynamic monitoring mechanism is constructed to track the quality of life of the target group and the implementation effect of the support policy, and use the feedback analysis method to identify problems in the policy implementation process and generate adjustment strategies, including: By collecting income data, employment data and educational resource data of the target group according to the monitoring cycle, and standardizing the collected data, the monitoring indicators and deviations of the group are calculated; Judging the quality of life and the effectiveness of policy implementation, including: Income bias, employment bias, and education coverage bias are calculated separately; Obtain the sum of weighted deviations of all target groups to obtain an overall execution score, and compare the overall execution score with a preset threshold. If the overall execution score is greater than or equal to the preset threshold, the overall execution status is good; if the overall execution score is less than the preset threshold, the overall execution status has problems; Based on the problems in the overall implementation status, feedback analysis algorithms are used to identify problem areas in policy implementation, and regression models are used to analyze the relationship between various policy indicators and deviations to determine the specific factors that affect policy implementation; By analyzing the deviation values, adjustment strategies are generated to optimize the allocation of supporting funds and resources.

8. The performance evaluation system for reservoir resettlement support funds is characterized by: The performance evaluation method for reservoir migrant support funds as claimed in any one of claims 1 to 7 comprises: a data collection module, the data collection module is used to obtain income data, employment status data, educational resource data of the target group, collect household registration information, community visit data and big data from external data sources; A target group determination module, which constructs a classification model for the target support group based on digital processing of household registration information, community visit data collection and big data analysis technology, and identifies the immigrant group affected by the reservoir construction according to the classification model and records it as the target group; A support fund performance indicator judgment module, which uses an income growth rate calculation algorithm, an employment rate statistical algorithm, and an education resource coverage rate calculation algorithm to calculate the income growth rate, employment rate, and education resource coverage rate of the target group, and builds a support fund performance evaluation model based on the calculation results, and evaluates whether the support fund performance indicators meet the preset standards according to the evaluation model; A dynamic adjustment module dynamically adjusts the allocation strategy of the support funds according to the evaluation results of the support fund performance evaluation model.

Images