Government affair service cross-platform affair handling process non-inductive evaluation method and device

Abstract

The invention discloses a government affair service cross-platform affair handling process non-inductive evaluation method and device, belongs to the technical field of government affair services, and aims to solve the technical problems that in the use process of software and hardware of a traditional government affair hall evaluation system, fairness cannot be guaranteed, tracing cannot be achieved, and the cost is high. According to the technical scheme, a sound pick-up is installed at a government affair service PC end, a dual-track automatic recording terminal is deployed, the sound pick-up and the dual-track automatic recording terminal are connected with a service acceptance system, recording is started while the service is handled, and recording corresponding to the currently handled service is uploaded; and the background server carries out AI identification on the recording through the large model, analyzes whether the service of the worker meets the requirement in the business handling process, carries out AI automatic scoring on the service, and archives the recording for tracing.

Description

Technical Field

[0001] The present invention relates to the technical field of government service, and in particular to a method and device for seamlessly evaluating the cross-platform government service process. Background Art

[0002] At present, the service evaluation in government service halls mainly uses evaluator equipment and programs, which prompt users to click on the evaluation to select good or bad reviews; this may cause the following problems: First, the evaluator equipment and the built-in evaluation system are relatively complex and costly, and it is difficult to modify them as the needs are upgraded during use; second, traditional evaluation can only record the evaluation results, but cannot trace whether the evaluation is fair and impartial, whether it is misleading, whether it is replaced by the evaluation, etc., and cannot distinguish between malicious and involuntary evaluations; third, with the rapid development of AI technology, traditional evaluation systems cannot effectively use large models for upgrades and optimization; fourth, with the promotion of domestic systems, some government service halls’ office computers have already implemented domestic systems, and traditional evaluation systems;

[0003] Traditional evaluation processing methods have brought many management difficulties. First, the existing evaluation recording method only records the evaluation results and cannot be traced or supervised. Second, a small number of people do not know how to operate or evaluate arbitrarily, or staff operate the evaluation on their behalf, which may lead to unfair and inaccurate evaluation results and make it impossible to distinguish malicious evaluations. Third, the traditional method has high hardware requirements, the system is complex and inflexible, and the upgrade cost is high.

[0004] In summary, the traditional government service hall evaluation system software and hardware cannot guarantee fairness and impartiality, cannot be traced, and has high costs during use, and has limited effect on improving the service level and fair and objective evaluation of the government service hall. Summary of the Invention

[0005] The technical task of the present invention is to provide a cross-platform government service process seamless evaluation method and device to solve the problems of fairness, justice, traceability and high cost in the use of software and hardware of traditional government hall evaluation systems.

[0006] The technical task of the present invention is achieved in the following way: a method for seamless evaluation of the cross-platform government service process, which is to install a microphone and deploy a dual-channel automatic recording terminal on the government service PC side, connect the microphone and the dual-channel automatic recording terminal with the business acceptance system, start recording while handling the business and upload the recording corresponding to the current business, and the background server uses a large model to perform AI recognition on the recording, analyze whether the service of the staff in the business handling process meets the requirements, and perform AI automatic scoring on this service, and archive the recording for retrospective use.

[0007] As a preferred option, the dual-channel automatic recording terminal uses the electron+vue cross-platform technology to achieve segmented recording, integrate the segmented recordings into a complete recording, and archive it; then upload the complete recording to the trained large model on the background server for recognition and scoring, and display the evaluation data and manage the data.

[0008] Preferably, the business acceptance system transmits the data of the service queue number, staff member's work number and business processing number to the dual-channel automatic recording terminal through the websocket interface.

[0009] Better yet, the trained large model uses deepseek to organize common conversations in government affairs based on the pre-trained language model, and customizes keywords for the scenario to train the model. After the trained model can accurately identify the content and semantics of the conversation, it will score the service process.

[0010] A cross-platform, seamless evaluation device for government service processes includes a government service PC terminal and a backend server, i.e., a business acceptance system. A microphone is installed on the government service PC terminal and a dual-channel automatic recording terminal is deployed. The microphone and dual-channel automatic recording terminal are connected to the business acceptance system.

[0011] Among them, the pickup is used to start recording while handling business and upload the recording corresponding to the current business;

[0012] The dual-channel automatic recording terminal is used for segmented recording, which is then integrated into a complete recording for archiving.

[0013] The backend server is used to deploy the trained large model, which performs AI recognition on recordings, analyzes whether the staff's service meets the requirements during the business process, automatically scores the service with AI, and archives the recordings for retrospective use.

[0014] As a preferred option, the dual-channel automatic recording terminal uses the electron+vue cross-platform technology to achieve segmented recording.

[0015] Preferably, the business acceptance system transmits the data of the service queue number, staff member's work number and business processing number to the dual-channel automatic recording terminal through the websocket interface.

[0016] As a preferred choice, the trained large model uses deepseek, organizes common conversations in government affairs based on the pre-trained language model, and trains the model based on customized keywords for the scenario. After the trained model can accurately identify the content and semantics of the conversation, it will score the service process.

[0017] More preferably, the working process of the device is as follows:

[0018] S1. The business acceptance system on the government service PC side calls numbers normally, and transmits the queue number, window number, staff member number and business information to the dual-channel automatic recording terminal through the websocket interface;

[0019] S2. The dual-channel automatic recording terminal receives the corresponding start-processing command, starts recording, and performs segmented recording every set time (e.g., 0.5 seconds), and uploads the recording to the background server;

[0020] S3. When the business acceptance system on the PC side of the government service completes the service normally, it operates the completion function. The business acceptance system transmits the completion signal and the completed queue number, window number, staff member number, and business information to the dual-channel automatic recording terminal through the websocket interface;

[0021] S4, the dual-channel automatic recording terminal receives the completion signal, stops recording, and transmits the completion signal and the number of fragments to the dual-channel automatic recording terminal;

[0022] S5. After receiving the completion signal, the dual-channel automatic recording terminal merges all received fragmented recordings to form a complete recording. After archiving the recording, it passes it to the trained large model for content recognition and scoring.

[0023] S6. The trained large model will archive the recognition content, provide a score, and store the evaluation record in the evaluation system;

[0024] S7: The backend server checks the evaluation results, conversation content, and recording content for retrospective use. The cross-platform, seamless evaluation method and device for government service processes of the present invention has the following advantages:

[0025] (1) The present invention realizes the senseless recording of the service process based on the hardware combination of a microphone and a two-channel automatic recording terminal, and performs AI evaluation through a large model; the present invention is introduced into the window service acceptance process of the government service hall for system transformation, which can be adapted to the domestic system or the existing desktop system to realize the senseless recording of the service dialogue, upload the background large model, perform AI intelligent analysis, and automatically evaluate the service process, so as to solve the problem that the service process of the government service hall staff is difficult to record and trace, and the problem that the traditional evaluator is high in cost and manual click evaluation cannot accurately reflect the service level, realize fair and objective evaluation and process traceability, and help improve the management and service level of the hall;

[0026] (2) Traditional evaluation systems usually include an evaluation tablet running Windows or Android, which is expensive. This application only requires a pickup that supports dual-channel recording, which greatly reduces the equipment investment cost and is low cost.

[0027] (3) The present invention realizes cross-platform recording through the transformation of the dual-channel automatic recording terminal, making conversation recording and evaluation easy and seamless;

[0028] (4) The present invention achieves intelligent and accurate dialogue content recognition and evaluation scoring through the introduction of AI large models;

[0029] (5) The present invention realizes the synthesis and archiving of recordings through the transformation of the background server, which is convenient for users to trace back. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] The present invention will be further described below with reference to the accompanying drawings.

[0031] Attachment Figure 1 A schematic diagram of the structure of a cross-platform, seamless evaluation device for government service processes;

[0032] Attachment Figure 2 A flowchart of the working process of a cross-platform seamless evaluation device for government service processes. DETAILED DESCRIPTION

[0033] The following detailed description is made of the method and device for seamlessly evaluating the cross-platform government service process of the present invention with reference to the accompanying drawings and specific embodiments.

[0034] Example 1:

[0035] This embodiment provides a method for seamlessly evaluating the cross-platform process of government services. The method is to install a microphone and deploy a dual-channel automatic recording terminal on the government service PC, connect the microphone and the dual-channel automatic recording terminal with the business acceptance system, start recording while handling the business and upload the recording corresponding to the current business. The background server uses a large model to perform AI recognition on the recording, analyze whether the service of the staff during the business handling process meets the requirements, and automatically score the service with AI, and archive the recording for retrospective use.

[0036] The dual-channel automatic recording terminal in this embodiment uses the electron+vue cross-platform technology to achieve fragmented recording, integrates the fragmented recordings into a complete recording, and archives it; then uploads the complete recording to the trained large model on the background server for recognition and scoring, and displays the evaluation data and manages the data.

[0037] The dual-channel automatic recording terminal in this embodiment provides a websocket interface, which is provided to the external business acceptance system to transmit data such as the service queue number, staff member number, and business processing number; that is, the business acceptance system transmits the data of the service queue number, staff member number, and business processing number to the dual-channel automatic recording terminal through the websocket interface.

[0038] The trained large model in this embodiment uses deepseek, organizes common government service conversations based on the pre-trained language model, and trains the model based on customized keywords for the scenario. After the trained model accurately recognizes the content and semantics of the conversation, it scores the service process.

[0039] Example 2:

[0040] As attached Figure 1 As shown, this embodiment provides a cross-platform, seamless evaluation device for government service processes. The device includes a government service PC terminal and a backend server, namely, a business acceptance system. A microphone is installed on the government service PC terminal and a dual-channel automatic recording terminal is deployed. The microphone and the dual-channel automatic recording terminal are connected to the business acceptance system.

[0041] Among them, the pickup is used to start recording while handling business and upload the recording corresponding to the current business;

[0042] The dual-channel automatic recording terminal is used for segmented recording, which is then integrated into a complete recording for archiving.

[0043] The backend server is used to deploy the trained large model, which performs AI recognition on recordings, analyzes whether the staff's service meets the requirements during the business process, automatically scores the service with AI, and archives the recordings for retrospective use.

[0044] The dual-channel automatic recording terminal in this embodiment implements segmented recording through the electron+vue cross-platform technology.

[0045] The business acceptance system in this embodiment transmits the data of the service queue number, staff member work number and business processing number to the dual-channel automatic recording terminal through the websocket interface.

[0046] The trained large model in this embodiment uses deepseek, organizes common government service conversations based on the pre-trained language model, and trains the model based on customized keywords for the scenario. After the trained model accurately recognizes the content and semantics of the conversation, it scores the service process.

[0047] As attached Figure 2 As shown, the working process of the device is as follows:

[0048] S1. The business acceptance system on the government service PC side calls numbers normally, and transmits the queue number, window number, staff member number and business information to the dual-channel automatic recording terminal through the websocket interface;

[0049] S2. The dual-channel automatic recording terminal receives the corresponding start-processing command, starts recording, and performs segmented recording every set time (e.g., 0.5 seconds), and uploads the recording to the background server;

[0050] S3. When the business acceptance system on the PC side of the government service completes the service normally, it operates the completion function. The business acceptance system transmits the completion signal and the completed queue number, window number, staff member number, and business information to the dual-channel automatic recording terminal through the websocket interface;

[0051] S4, the dual-channel automatic recording terminal receives the completion signal, stops recording, and transmits the completion signal and the number of fragments to the dual-channel automatic recording terminal;

[0052] S5. After receiving the completion signal, the dual-channel automatic recording terminal merges all received fragmented recordings to form a complete recording. After archiving the recording, it passes it to the trained large model for content recognition and scoring.

[0053] S6. The trained large model will archive the recognition content, provide a score, and store the evaluation record in the evaluation system;

[0054] S7. The backend server checks the evaluation results, conversation content, and recording content for traceability.

[0055] Example 3:

[0056] Taking the actual situation of a project site as an example, the implementation methods are mainly as follows:

[0057] First, deploy a large model on the backend server and train it with common conversations and keywords. Then deploy a server-side program to provide an interface for uploading recordings and storage space for recordings.

[0058] Second, install and debug the hardware pickup on site, install the dual-channel automatic recording terminal, and debug and operate the recording function;

[0059] Third, the acceptance and calling system is modified to call the websocket interface of the recording terminal to transmit information when calling and completing the process.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or replace some or all of the technical features therein with equivalents. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A cross-platform process evaluation method for government services, characterized by: This method is to install a microphone and deploy a dual-channel automatic recording terminal on the government service PC, connect the microphone and the dual-channel automatic recording terminal with the business acceptance system, start recording while handling the business and upload the recording corresponding to the current business. The background server uses a large model to perform AI recognition on the recording, analyze whether the service of the staff meets the requirements during the business handling process, and automatically score the service with AI, and archive the recording for retrospective use.

2. The method for seamlessly evaluating the cross-platform government service process according to claim 1 is characterized in that: The dual-channel automatic recording terminal uses the electron+vue cross-platform technology to achieve segmented recording, integrates the segmented recordings into a complete recording, and archives it; then uploads the complete recording to the trained large model on the background server for recognition and scoring, and displays the evaluation data and manages the data.

3. The method for seamlessly evaluating the cross-platform government service process according to claim 1 is characterized in that: The business acceptance system transmits the data of service queue number, staff member work number and business processing number to the dual-channel automatic recording terminal through the websocket interface.

4. The method for seamlessly evaluating the cross-platform government service process according to any one of claims 1 to 3, characterized in that: The trained large model uses deepseek, organizes common conversations in government affairs based on the pre-trained language model, and trains the model with customized keywords based on the scenario. After the trained model can accurately identify the content and semantics of the conversation, it scores the service process.

5. A cross-platform government service process seamless evaluation device, characterized by: The device includes a government service PC terminal and a backend server, namely a business acceptance system. A microphone is installed on the government service PC terminal and a dual-channel automatic recording terminal is deployed. The microphone and the dual-channel automatic recording terminal are connected to the business acceptance system. Among them, the pickup is used to start recording while handling business and upload the recording corresponding to the current business; The dual-channel automatic recording terminal is used for segmented recording, which is then integrated into a complete recording for archiving. The backend server is used to deploy the trained large model, which performs AI recognition on recordings, analyzes whether the staff's service meets the requirements during the business process, automatically scores the service with AI, and archives the recordings for retrospective use.

6. The cross-platform process-free evaluation device for government services according to claim 5 is characterized in that: The dual-channel automatic recording terminal uses the electron+vue cross-platform technology to achieve segmented recording.

7. The cross-platform process-free evaluation device for government services according to claim 5 is characterized in that: The business acceptance system transmits the data of service queue number, staff member work number and business processing number to the dual-channel automatic recording terminal through the websocket interface.

8. The cross-platform process-free evaluation device for government services according to claim 5 is characterized in that: The trained large model uses deepseek, organizes common conversations in government affairs based on the pre-trained language model, and trains the model with customized keywords based on the scenario. After the trained model can accurately identify the content and semantics of the conversation, it scores the service process.

9. The cross-platform government service process seamless evaluation device according to any one of claims 5 to 8, characterized in that: The working process of the device is as follows: S1. The business acceptance system on the government service PC side calls numbers normally, and transmits the queue number, window number, staff member number and business information to the dual-channel automatic recording terminal through the websocket interface; S2. The dual-channel automatic recording terminal receives the corresponding start-processing command, starts recording, and performs segmented recording at set intervals, and uploads the recording to the backend server; S3. When the business acceptance system on the PC side of the government service completes the service normally, it operates the completion function. The business acceptance system transmits the completion signal and the completed queue number, window number, staff member number, and business information to the dual-channel automatic recording terminal through the websocket interface; S4, the dual-channel automatic recording terminal receives the completion signal, stops recording, and transmits the completion signal and the number of fragments to the dual-channel automatic recording terminal; S5. After receiving the completion signal, the dual-channel automatic recording terminal merges all received fragmented recordings to form a complete recording. After archiving the recording, it passes it to the trained large model for content recognition and scoring. S6. The trained large model will archive the recognition content, provide a score, and store the evaluation record in the evaluation system; S7. The backend server checks the evaluation results, conversation content, and recording content for traceability.

Images