A government affair service optimization method and system based on big data analysis

Abstract

The application is suitable for the technical field of service optimization, and particularly relates to a government service optimization method and system based on big data analysis. The method comprises the following steps: setting a plurality of source data nodes, wherein the plurality of source data nodes at least comprise a household registration node, a social security node and a tax node; collecting a single service event of each source data node; obtaining feedback data of evaluation indexes, wherein the evaluation indexes at least comprise an approval pass rate, a repeated handling frequency and a complaint rate; building an evaluation channel; obtaining a satisfaction score corresponding to the feedback data in the single service event; generating a training set; and training a pre-constructed government service performance evaluation model. The application can intuitively display the evaluation deficiencies of government data by matching and analyzing the output results corresponding to each evaluation index with a preset fluctuation range, thereby enhancing the pertinence of problem feedback and significantly improving the fine level and problem discovery capability of government service performance evaluation.

Description

Technical Field

[0001] This invention relates to the field of service optimization technology, and in particular to a method and system for optimizing government services based on big data analysis. Background Technology

[0002] With the continuous advancement of information technology development, government services are gradually transforming from the traditional offline processing model to an integrated online and offline service model. Management departments have aggregated a large amount of business processing data, user behavior data, and multi-source heterogeneous auxiliary data through government service platforms, and are using big data analytics to identify service deficiencies to improve service efficiency. However, the existing government service system still has certain limitations in data utilization, mainly reflected in data fragmentation, inconsistent standards, and insufficient cross-departmental collaboration. For example, due to data sensitivity, different departments cannot display certain sensitive business process data, leading to discrepancies in service satisfaction evaluations.

[0003] Therefore, "how to assess service quality when there are access restrictions on cross-departmental data" is the technical problem that this invention aims to solve. Summary of the Invention

[0004] The purpose of this invention is to provide a method and system for optimizing government services based on big data analysis, in order to solve the problem mentioned in the background art of "how to evaluate service quality when there are access restrictions on cross-departmental data".

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A method for optimizing government services based on big data analytics, the method comprising:

[0007] Set up multi-source data nodes, wherein the multi-source data nodes include at least: household registration node, social security node and tax node, collect single service events of each multi-source data node, and obtain feedback data of evaluation indicators, wherein the evaluation indicators include at least: approval rate, number of repeated processing and complaint rate.

[0008] An evaluation channel is established to obtain the satisfaction score corresponding to the feedback data in a single service event, a training set is generated, and a pre-built government service performance evaluation model is trained. Each multi-source data node corresponds to a government service performance evaluation model. The model parameters of each government service performance evaluation model are extracted, the weight values ​​corresponding to each multi-source data node are set, and a weighted average is performed to obtain global parameters. The global parameters are written into the initialized government service performance evaluation model to generate a global model.

[0009] After collecting the real-time value of a single service event, the real-time value is input into the global model, and an evaluation score is output. When the evaluation score is less than the threshold, the real-time value is input again into the corresponding government service performance evaluation model to obtain the output result. Each multi-source data node corresponds to an output result. A fluctuation range corresponding to the evaluation index is created. The output result and the fluctuation range are compared to define anomalies. The anomalies and their corresponding fluctuation ranges are written into a preset template to generate an optimization report, which is then sent to a preset terminal.

[0010] Furthermore, the step of setting up multi-source data nodes, wherein the multi-source data nodes include at least: household registration nodes, social security nodes, and tax nodes, includes:

[0011] Establish the correspondence between multi-source data nodes, evaluation indicators and feedback data, and generate service evaluation logs;

[0012] Record the generation time of each service evaluation log, sort them in chronological order, and build a log set.

[0013] Furthermore, the step of collecting single service events from each multi-source data node and obtaining feedback data for evaluation metrics includes:

[0014] Determine the benchmark value for each evaluation indicator. When the feedback data is less than the benchmark value, add the corresponding evaluation indicator to the anomaly list.

[0015] Furthermore, the steps of establishing an evaluation channel, obtaining satisfaction scores corresponding to feedback data in a single service event, and generating a training set include:

[0016] Set the evaluation time points for satisfaction ratings, and plot a trend graph with the evaluation time points as the horizontal axis and the satisfaction ratings at the corresponding times as the vertical axis.

[0017] Iterate through the inflection points in the trend chart, define the corresponding single service events as reference events, and forward them to the preset terminals.

[0018] Furthermore, the steps of setting the weight value corresponding to each multi-source data node, performing a weighted average to obtain global parameters, and writing the global parameters into the initialized government service performance evaluation model to generate the global model include:

[0019] The global model is updated according to a preset time step, and the global parameters are versioned and stored to obtain several versions.

[0020] All versions are aggregated, a version chain is generated, and a rollback mechanism is embedded.

[0021] Furthermore, the steps of comparing the output results and fluctuation ranges, defining anomalies, writing the anomalies and their corresponding fluctuation ranges into a preset template, generating an optimization report, and sending it to a preset terminal include:

[0022] The factors influencing the configuration threshold include at least: time and service load, and the threshold is dynamically adjusted using the factors influencing the configuration threshold.

[0023] Create an update window for the threshold and draw a comparison table.

[0024] Furthermore, the system includes:

[0025] The acquisition module is used to set up multi-source data nodes, wherein the multi-source data nodes include at least: household registration node, social security node and tax node, collect single service events of each multi-source data node, and obtain feedback data of evaluation indicators, wherein the evaluation indicators include at least: approval rate, number of repeated processing times and complaint rate.

[0026] The writing module is used to build an evaluation channel, obtain the satisfaction score corresponding to the feedback data in a single service event, generate a training set, and train a pre-built government service performance evaluation model. Each multi-source data node corresponds to a government service performance evaluation model. The model parameters of each government service performance evaluation model are extracted, the weight values ​​corresponding to each multi-source data node are set, and a weighted average is performed to obtain global parameters. The global parameters are written into the initialized government service performance evaluation model to generate a global model.

[0027] The distribution module is used to input the real-time value of a single service event into the global model after it is collected, and output an evaluation score. When the evaluation score is less than a threshold, the real-time value is input into the corresponding government service performance evaluation model again to obtain an output result. Each multi-source data node corresponds to an output result. A fluctuation range corresponding to the evaluation index is created. The output result and the fluctuation range are compared to define anomalies. The anomalies and their corresponding fluctuation ranges are written into a preset template to generate an optimization report and distribute it to a preset terminal.

[0028] Furthermore, the acquisition module includes:

[0029] Establish a unit to establish the correspondence between multi-source data nodes, evaluation indicators and feedback data, and generate service evaluation logs;

[0030] The recording unit is used to record the generation time of each service evaluation log and sort them in chronological order to form a log set;

[0031] An addition unit is used to determine the benchmark value corresponding to each evaluation indicator. When the feedback data is less than the benchmark value, the corresponding evaluation indicator is added to the anomaly list.

[0032] Furthermore, the writing module includes:

[0033] The setting unit is used to set the evaluation time point for the satisfaction score. The evaluation time point is used as the horizontal axis and the satisfaction score at the corresponding time point is used as the vertical axis to draw a trend graph.

[0034] The traversal unit is used to traverse the inflection point part of the trend chart, define the corresponding single service event as a reference event, and forward it to the preset terminal.

[0035] The unit is used to update the global model according to a preset time step and to store the global parameters in a versioned manner to obtain several versions.

[0036] The embedded unit is used to aggregate all versions, generate a version chain, and embed a rollback mechanism.

[0037] Furthermore, the distribution module includes:

[0038] A configuration unit is used to configure the influencing factors of a threshold, wherein the influencing factors include at least: time and service load, and to dynamically adjust the threshold using the influencing factors;

[0039] The drawing unit is used to create an update window for the threshold and draw a lookup table.

[0040] Compared with the prior art, the beneficial effects of the present invention are:

[0041] By setting up multi-source data nodes and constructing corresponding government service performance evaluation models for each, data from different business areas such as household registration, social security, and taxation can be evaluated and analyzed in a targeted manner within their respective models, greatly improving the accuracy of the analysis. By using the weighted fusion of parameters from each model to generate a unified global model, collaborative modeling and comprehensive evaluation of cross-departmental government service data can be achieved. While ensuring data privacy, the evaluation of the service process can be realized by sharing model parameters, significantly improving the overall consistency and accuracy of the evaluation results. By matching the output results corresponding to each evaluation indicator with preset fluctuation ranges, the shortcomings in the evaluation of government data can be intuitively revealed, enhancing the targeting of problem feedback and significantly improving the refinement level and problem detection capability of government service performance evaluation. Attached Figure Description

[0042] Figure 1 A flowchart illustrating the government service optimization method based on big data analysis provided in this embodiment of the invention;

[0043] Figure 2 A first sub-process flowchart of the government service optimization method based on big data analysis provided in an embodiment of the present invention;

[0044] Figure 3 The second sub-process flowchart of the government service optimization method based on big data analysis provided in the embodiments of the present invention;

[0045] Figure 4 The third sub-process flowchart of the government service optimization method based on big data analysis provided in the embodiments of the present invention;

[0046] Figure 5 A block diagram illustrating the composition of a government service optimization system based on big data analytics, as provided in an embodiment of the present invention.

[0047] Figure 6 A block diagram illustrating the composition of the acquisition module in the government service optimization system based on big data analysis provided in this embodiment of the invention;

[0048] Figure 7 A block diagram of the writing module in the government service optimization system based on big data analysis provided in an embodiment of the present invention;

[0049] Figure 8 This is a block diagram of the distribution module in the government service optimization system based on big data analysis provided in an embodiment of the present invention. Detailed Implementation

[0050] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0051] In Example 1, Figure 1 The implementation flow of the government service optimization method based on big data analysis provided in this embodiment of the invention is illustrated below:

[0052] S100: Set up multi-source data nodes, wherein the multi-source data nodes include at least: household registration node, social security node and tax node, collect single service events of each multi-source data node, and obtain feedback data of evaluation indicators, wherein the evaluation indicators include at least: approval rate, number of repeated processing times and complaint rate.

[0053] Multi-source data nodes are set up, including household registration, social security, and tax nodes, to respectively carry out service data collection and processing functions in their respective government business areas. A single service event is defined, for example, a single service event is the complete business process from application submission, review and processing to result feedback within one application cycle. The entire process business data corresponding to this single service event is acquired, along with specific user data under each evaluation indicator, i.e., feedback data. The evaluation indicators are user evaluations, and the feedback data are the specific numerical values ​​of these evaluations. Evaluation indicators include: approval rate, number of repeat applications, and complaint rate, etc. The advantage of this approach is that, through the unified collection and standardized processing of the above evaluation indicators, a quantitative expression of the service quality of different multi-source data nodes is achieved, providing a data foundation for subsequent multi-source data fusion analysis and evaluation.

[0054] S200: Establish an evaluation channel, obtain the satisfaction score corresponding to the feedback data in a single service event, generate a training set, and train a pre-built government service performance evaluation model. Each multi-source data node corresponds to a government service performance evaluation model. Extract the model parameters of each government service performance evaluation model, set the weight value corresponding to each multi-source data node, and perform a weighted average to obtain global parameters. Write the global parameters into the initialized government service performance evaluation model to generate a global model.

[0055] An evaluation channel is established, including online evaluation portals, mobile feedback interfaces, and offline terminal interactions, to obtain satisfaction scores corresponding to each service event. These satisfaction scores are specified by the users; it's important to note that "users" here refers to applicants, members of the public, or service recipients of government services. Each individual service event corresponds to a satisfaction score. The collected feedback data is cleaned, denoised, and standardized. Evaluation indicators such as approval rate, number of repeat applications, and complaint rate are then characterized according to preset field rules to construct a training set. A government service performance evaluation model is constructed using deep learning algorithms. This model analyzes and determines the satisfaction score for each service event based on the feedback data. The training set is input into the model, and the model parameters are optimized and adjusted through iterative training. This allows the model to learn the mapping relationship between multi-source data features and satisfaction scores, with each multi-source data node corresponding to a government service performance evaluation model.

[0056] After training, the model parameters corresponding to each multi-source data node are extracted. These parameters include, but are not limited to, weight coefficients, bias terms, and feature mapping parameters. Based on the importance, data integrity, historical stability, or business proportion of each multi-source data node in the government service process, each multi-source data node corresponds to a set of model parameters. A corresponding weight value is configured for each multi-source data node, where the weight value reflects the difference in contribution of different nodes in the overall evaluation system. The model parameters corresponding to each node are then weighted and averaged, i.e., the model parameters of each node are multiplied and summed with their corresponding weight values ​​to calculate the average value. This average value is the global parameter. The global parameter is written into the initialized government service performance evaluation model, and the parameters of the initialized model are updated to obtain the global model.

[0057] S300: After collecting the real-time value of a single service event, the real-time value is input into the global model, and an evaluation score is output. When the evaluation score is less than the threshold, the real-time value is input into the corresponding government service performance evaluation model again to obtain the output result. Each multi-source data node corresponds to an output result. A fluctuation range corresponding to the evaluation index is created. The output result and the fluctuation range are compared to define anomalies. The anomalies and their corresponding fluctuation ranges are written into a preset template to generate an optimization report and send it to a preset terminal.

[0058] During real-time monitoring, after collecting real-time business data corresponding to a single service event, the real-time value is input into the global model. The global model outputs the corresponding satisfaction score, i.e., the evaluation score, which characterizes the service quality level of the current service event. The evaluation score is compared with a preset threshold. When the evaluation score is less than the threshold, it is determined that the current service event may have a service anomaly, triggering secondary analysis. The real-time value is mapped to the corresponding multi-source data node according to its business category, and the real-time value is input again into the government service performance evaluation model corresponding to the multi-source data node for local refined analysis. The advantage of this method is that it can not only achieve rapid assessment of user satisfaction, but also find service defects, greatly improving the accuracy of anomaly identification and the precision of problem localization.

[0059] For each evaluation indicator, a corresponding fluctuation range is constructed, determined based on historical data, industry benchmark values, or model prediction intervals. After obtaining the output results corresponding to the multi-source data nodes, each output result is compared and analyzed with the fluctuation range of its corresponding evaluation indicator to determine whether the output result falls within the corresponding fluctuation range. When an output result exceeds the fluctuation range, it is marked as an anomaly. The anomaly is written into a preset template to generate an evaluation report, which is then distributed to a preset terminal, which can be a government service management personnel terminal.

[0060] In Example 2, Figure 2 The diagram illustrates the first sub-process flowchart of the government service optimization method based on big data analysis provided in this embodiment of the invention. The following details the step of setting up multi-source data nodes, wherein the multi-source data nodes include at least: household registration nodes, social security nodes, and tax nodes, as follows:

[0061] S101: Establish the correspondence between multi-source data nodes, evaluation indicators and feedback data, and generate service evaluation logs.

[0062] Determine the evaluation metrics and feedback data corresponding to each multi-source data node, establish the correspondence between the three, summarize them, and generate a service evaluation log.

[0063] S102: Record the generation time of each service evaluation log, sort them in chronological order, and build a log set.

[0064] Determine the generation time of each service evaluation log, sort the service evaluation logs in chronological order, and build a log set, which is the collection of service evaluation logs sorted by time sequence.

[0065] In Example 3, Figure 2 The first sub-process flowchart of the government service optimization method based on big data analysis provided in this embodiment of the invention is shown. The following details the steps of collecting single service events from each multi-source data node and obtaining feedback data of evaluation indicators:

[0066] S103: Determine the benchmark value corresponding to each evaluation indicator. When the feedback data is less than the benchmark value, add the corresponding evaluation indicator to the anomaly list.

[0067] A corresponding benchmark value is set for each evaluation indicator. The benchmark value is also the assessment threshold, which is the minimum value to determine whether the standard is met. When the feedback data is less than the benchmark value, it indicates that there is an abnormality in the corresponding evaluation indicator, and the evaluation indicator is defined as an abnormal item.

[0068] In Example 4, Figure 3 The second sub-process flowchart of the government service optimization method and system based on big data analysis provided in this embodiment of the invention is shown. The following details the steps of building an evaluation channel, obtaining the satisfaction score corresponding to the feedback data in a single service event, and generating a training set:

[0069] S201: Set the evaluation time point for the satisfaction rating, and plot the trend graph with the evaluation time point as the horizontal axis and the satisfaction rating at the corresponding time point as the vertical axis.

[0070] Determine the evaluation time points for satisfaction ratings. These time points can be uniformly calibrated based on the occurrence time of a single service event, processing progress milestones, or preset time intervals. Record the satisfaction score corresponding to each evaluation time point. Plot the evaluation time point as the horizontal axis and the corresponding satisfaction score as the vertical axis, map and connect the data from each time point to obtain a trend graph. This trend graph reflects the dynamic changes in satisfaction over time or as the service process progresses.

[0071] S202: Traverse the inflection point section of the trend chart, define the corresponding single service event as a reference event, and forward it to the preset terminal.

[0072] The trend chart is analyzed traversally. By calculating the slope of change between adjacent data points, locations where significant changes in the trend occur are identified and defined as inflection point regions. These inflection point regions represent key nodes where satisfaction scores shift from rising to falling, or from falling to rising. A portion of the individual service events corresponding to the inflection point regions is defined as reference events. These reference events are used for subsequent analysis of the causes of service process anomalies, comparative evaluation, or model calibration. The reference events are then sent to a preset terminal.

[0073] In Example 5, Figure 3 The second sub-process flowchart of the government service optimization method based on big data analysis provided in this embodiment of the invention is shown. The following details the steps of setting the weight value corresponding to each multi-source data node, performing a weighted average to obtain global parameters, and writing the global parameters into the initialized government service performance evaluation model to generate the global model:

[0074] S203: Update the global model according to the preset time step, and store the global parameters in a versioned manner to obtain several versions.

[0075] The global parameters are updated according to the preset time step, and the global parameters are stored in a versioned manner to obtain several versions.

[0076] S204: Aggregate all versions, generate a version chain, and embed a rollback mechanism.

[0077] All versions are linked in chronological order to generate a version chain, which is a collection of versions sorted by creation time. A rollback mechanism is embedded in the version chain, which means that when a new version has an anomaly, it can be quickly restored to a historical stable version.

[0078] In Example 6, Figure 4The diagram illustrates the third sub-process flowchart of the government service optimization method based on big data analysis provided in this embodiment of the invention. The following details the steps of comparing the output results and fluctuation ranges, defining anomalies, writing the anomalies and their corresponding fluctuation ranges into a preset template, generating an optimization report, and sending it to a preset terminal:

[0079] S301: Factors affecting the configuration threshold, wherein the factors affecting the configuration threshold include at least: time and service load, and the threshold is dynamically adjusted using the factors affecting the configuration threshold.

[0080] Identify the factors that may affect the threshold, i.e., the influencing factors. These factors include time and service load. Use these factors to adjust the threshold.

[0081] S302: Create an update window for the threshold and draw a lookup table.

[0082] After the threshold is updated, record the change process of the threshold, determine the time window corresponding to the change process, i.e. the update window, integrate the update window and the corresponding threshold, and draw a comparison table.

[0083] Figure 5 The diagram illustrates the composition of a government service optimization system based on big data analytics provided in an embodiment of the present invention. The government service optimization system 1 based on big data analytics includes:

[0084] The acquisition module 11 is used to set up multi-source data nodes, wherein the multi-source data nodes include at least: household registration node, social security node and tax node, collect single service events of each multi-source data node, and obtain feedback data of evaluation indicators, wherein the evaluation indicators include at least: approval rate, number of repeated processing times and complaint rate.

[0085] The writing module 12 is used to build an evaluation channel, obtain the satisfaction score corresponding to the feedback data in a single service event, generate a training set, and train a pre-built government service performance evaluation model. Each multi-source data node corresponds to a government service performance evaluation model. The model parameters of each government service performance evaluation model are extracted, the weight values ​​corresponding to each multi-source data node are set, and a weighted average is performed to obtain global parameters. The global parameters are written into the initialized government service performance evaluation model to generate a global model.

[0086] The distribution module 13 is used to input the real-time value of a single service event into the global model after it is collected, and output an evaluation score. When the evaluation score is less than a threshold, the real-time value is input into the corresponding government service performance evaluation model again to obtain an output result. Each multi-source data node corresponds to an output result. A fluctuation range corresponding to the evaluation index is created. The output result and the fluctuation range are compared to define anomalies. The anomalies and the corresponding fluctuation range are written into a preset template to generate an optimization report and distribute it to a preset terminal.

[0087] Figure 6 This diagram illustrates the composition of the acquisition module 11 in the government service optimization system based on big data analysis provided in an embodiment of the present invention. The acquisition module 11 includes:

[0088] Establish unit 111 to establish the correspondence between multi-source data nodes, evaluation indicators and feedback data, and generate service evaluation logs;

[0089] Recording unit 112 is used to record the generation time of each service evaluation log and sort them in chronological order to form a log set;

[0090] Adding unit 113 is used to determine the benchmark value corresponding to each evaluation indicator. When the feedback data is less than the benchmark value, the corresponding evaluation indicator is added to the anomaly list.

[0091] Figure 7 This diagram illustrates the composition of the writing module 12 in the government service optimization system based on big data analysis provided in an embodiment of the present invention. The writing module 12 includes:

[0092] Setting unit 121 is used to set the evaluation time point of the satisfaction score, and plot the trend graph with the evaluation time point as the horizontal axis and the satisfaction score at the corresponding time point as the vertical axis.

[0093] Traversal unit 122 is used to traverse the inflection point part of the trend chart, define the corresponding single service event as a reference event, and forward it to the preset terminal.

[0094] Unit 123 is used to update the global model according to a preset time step and to store the global parameters in a versioned manner to obtain several versions.

[0095] Embedded unit 124 is used to aggregate all versions, generate a version chain, and embed a rollback mechanism.

[0096] Figure 8 This diagram illustrates the composition of the distribution module 13 in the government service optimization system based on big data analysis provided in an embodiment of the present invention. The distribution module 13 includes:

[0097] Configuration unit 131 is used to configure the influencing factors of the threshold, wherein the influencing factors include at least: time and service load, and to dynamically adjust the threshold using the influencing factors;

[0098] Drawing unit 132 is used to create an update window for the threshold and draw a lookup table.

[0099] The acquisition module 11 is mainly used to complete step S100, the writing module 12 is mainly used to complete step S200, and the sending module 13 is mainly used to complete step S300.

[0100] The establishment unit 111 is mainly used to complete step S101, the recording unit 112 is mainly used to complete step S102, and the addition unit 113 is mainly used to complete step S103.

[0101] The setting unit 121 is mainly used to complete step S201, the traversal unit 122 is mainly used to complete step S202, the obtaining unit 123 is mainly used to complete step S203, and the embedding unit 124 is mainly used to complete step S204.

[0102] The configuration unit 131 is mainly used to complete step S301, and the drawing unit 132 is mainly used to complete step S302.

[0103] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0104] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0105] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for optimizing government services based on big data analysis, characterized in that, The method includes: Set up multi-source data nodes, wherein the multi-source data nodes include at least: household registration node, social security node and tax node, collect single service events of each multi-source data node, and obtain feedback data of evaluation indicators, wherein the evaluation indicators include at least: approval rate, number of repeated processing and complaint rate. An evaluation channel is established to obtain the satisfaction score corresponding to the feedback data in a single service event, a training set is generated, and a pre-built government service performance evaluation model is trained. Each multi-source data node corresponds to a government service performance evaluation model. The model parameters of each government service performance evaluation model are extracted, the weight values ​​corresponding to each multi-source data node are set, and a weighted average is performed to obtain global parameters. The global parameters are written into the initialized government service performance evaluation model to generate a global model. After collecting the real-time value of a single service event, the real-time value is input into the global model, and an evaluation score is output. When the evaluation score is less than the threshold, the real-time value is input again into the corresponding government service performance evaluation model to obtain the output result. Each multi-source data node corresponds to an output result. A fluctuation range corresponding to the evaluation index is created. The output result and the fluctuation range are compared to define anomalies. The anomalies and their corresponding fluctuation ranges are written into a preset template to generate an optimization report, which is then sent to a preset terminal.

2. The method for optimizing government services based on big data analysis according to claim 1, characterized in that, The step of setting up multi-source data nodes, wherein the multi-source data nodes include at least: household registration nodes, social security nodes, and tax nodes, includes: Establish the correspondence between multi-source data nodes, evaluation indicators and feedback data, and generate service evaluation logs; Record the generation time of each service evaluation log, sort them in chronological order, and build a log set.

3. The method for optimizing government services based on big data analysis according to claim 2, characterized in that, The steps of collecting single service events from each multi-source data node and obtaining feedback data for evaluation metrics include: Determine the benchmark value for each evaluation indicator. When the feedback data is less than the benchmark value, add the corresponding evaluation indicator to the anomaly list.

4. The method for optimizing government services based on big data analysis according to claim 1, characterized in that, The steps of establishing an evaluation channel, obtaining satisfaction scores corresponding to feedback data in a single service event, and generating a training set include: Set the evaluation time points for satisfaction ratings, and plot a trend graph with the evaluation time points as the horizontal axis and the satisfaction ratings at the corresponding times as the vertical axis. Iterate through the inflection points in the trend chart, define the corresponding single service events as reference events, and forward them to the preset terminals.

5. The method for optimizing government services based on big data analysis according to claim 1, characterized in that, The steps of setting the weight value corresponding to each multi-source data node, performing a weighted average to obtain global parameters, and writing the global parameters into the initialized government service performance evaluation model to generate the global model include: The global model is updated according to a preset time step, and the global parameters are versioned and stored to obtain several versions. All versions are aggregated, a version chain is generated, and a rollback mechanism is embedded.

6. The method for optimizing government services based on big data analysis according to claim 1, characterized in that, The steps of comparing the output results and fluctuation ranges, defining anomalies, writing the anomalies and their corresponding fluctuation ranges into a preset template, generating an optimization report, and sending it to a preset terminal include: The factors influencing the configuration threshold include at least: time and service load, and the threshold is dynamically adjusted using these factors. Create an update window for the threshold and draw a comparison table.

7. A government service optimization system based on big data analysis, characterized in that, The system includes: The acquisition module is used to set up multi-source data nodes, wherein the multi-source data nodes include at least: household registration node, social security node and tax node, collect single service events of each multi-source data node, and obtain feedback data of evaluation indicators, wherein the evaluation indicators include at least: approval rate, number of repeated processing times and complaint rate. The writing module is used to build an evaluation channel, obtain the satisfaction score corresponding to the feedback data in a single service event, generate a training set, and train a pre-built government service performance evaluation model. Each multi-source data node corresponds to a government service performance evaluation model. The model parameters of each government service performance evaluation model are extracted, the weight values ​​corresponding to each multi-source data node are set, and a weighted average is performed to obtain global parameters. The global parameters are written into the initialized government service performance evaluation model to generate a global model. The distribution module is used to input the real-time value of a single service event into the global model after it is collected, and output an evaluation score. When the evaluation score is less than a threshold, the real-time value is input into the corresponding government service performance evaluation model again to obtain an output result. Each multi-source data node corresponds to an output result. A fluctuation range corresponding to the evaluation index is created. The output result and the fluctuation range are compared to define anomalies. The anomalies and their corresponding fluctuation ranges are written into a preset template to generate an optimization report and distribute it to a preset terminal.

8. The government service optimization system based on big data analysis according to claim 7, characterized in that, The acquisition module includes: Establish a unit to establish the correspondence between multi-source data nodes, evaluation indicators and feedback data, and generate service evaluation logs; The recording unit is used to record the generation time of each service evaluation log and sort them in chronological order to form a log set; An addition unit is used to determine the benchmark value corresponding to each evaluation indicator. When the feedback data is less than the benchmark value, the corresponding evaluation indicator is added to the anomaly list.

9. The government service optimization system based on big data analysis according to claim 7, characterized in that, The writing module includes: The setting unit is used to set the evaluation time point for the satisfaction score. The evaluation time point is used as the horizontal axis and the satisfaction score at the corresponding time point is used as the vertical axis to draw a trend graph. The traversal unit is used to traverse the inflection point part of the trend chart, define the corresponding single service event as a reference event, and forward it to the preset terminal. The unit is used to update the global model according to a preset time step and to store the global parameters in a versioned manner to obtain several versions. The embedded unit is used to aggregate all versions, generate a version chain, and embed a rollback mechanism.

10. The government service optimization system based on big data analysis according to claim 7, characterized in that, The distribution module includes: A configuration unit is used to configure the influencing factors of a threshold, wherein the influencing factors include at least: time and service load, and to dynamically adjust the threshold using the influencing factors; The drawing unit is used to create an update window for the threshold and draw a lookup table.

Images