User feedback intelligent processing method and device, electronic equipment and readable storage medium

Abstract

The invention provides a user feedback intelligent processing method and device, electronic equipment and a readable storage medium, and the method comprises the steps: obtaining user feedback data which comprises current feedback data and feedback association data, the feedback association data comprises at least one of user attribute data, user order data, user operation data, historical interaction data and industry news data; performing data preprocessing and semantic analysis on the user feedback data to generate analysis data; analyzing the analysis data by using a problem analysis model to generate a problem report; and generating a problem work order according to the problem report, and sending the problem work order to a customer service terminal. According to the technical scheme provided by one or more embodiments, the client feedback problem can be efficiently and accurately processed, and the client service quality and the user satisfaction degree are improved.

Description

Technical Field

[0001] The present disclosure relates to the field of data processing technology, and in particular to a method, device, electronic device, and readable storage medium for intelligently processing user feedback. Background Art

[0002] With the rapid development of the online ride-hailing industry, whether or not customer feedback can be effectively handled has become an important basis for measuring the service quality of online ride-hailing companies and optimizing their operating strategies.

[0003] However, traditional customer relationship management (SCRM) solutions in the ride-hailing industry have numerous shortcomings. For one thing, manual customer service responses to customer feedback are inefficient and lack accuracy and consistency. Furthermore, traditional customer feedback processing methods often fail to capture customers' true needs and emotional tendencies, making it difficult to continuously improve customer service quality. Summary of the Invention

[0004] In view of this, one or more embodiments of the present disclosure provide a method, device, electronic device and readable storage medium for intelligent processing of user feedback, which can efficiently and accurately handle customer feedback issues and improve customer service quality and user satisfaction.

[0005] On the one hand, the present disclosure provides a method for intelligently processing user feedback, which is used in the online car-hailing industry. The method includes: obtaining user feedback data, the user feedback data including current feedback data and feedback-related data, the feedback-related data including at least one of user attribute data, user order data, user operation data, historical interaction data, and industry news data; performing data preprocessing and semantic parsing on the user feedback data to generate parsed data; using a problem analysis model to analyze the parsed data to generate a problem report; generating a problem work order based on the problem report, and sending the problem work order to a customer service terminal.

[0006] On the other hand, the present disclosure also provides a user feedback intelligent processing device, which is used in the online car-hailing industry, and the device includes: a data acquisition unit, used to obtain user feedback data, the user feedback data includes current feedback data and feedback-related data, and the feedback-related data includes at least one of user attribute data, user order data, user operation data, historical interaction data, and industry news data; a data parsing unit, used to perform data preprocessing and semantic parsing on the user feedback data to generate parsed data; a problem analysis unit, used to analyze the parsed data using a problem analysis model to generate a problem report; and a work order processing unit, used to generate a problem work order based on the problem report, and send the problem work order to a customer service terminal.

[0007] On the other hand, the present disclosure further provides an electronic device, which includes a memory and a processor, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, the above-mentioned user feedback intelligent processing method is implemented.

[0008] On the other hand, the present disclosure further provides a computer-readable storage medium, which is used to store a computer program. When the computer program is executed by a processor, the above-mentioned user feedback intelligent processing method is implemented.

[0009] The technical solution provided by one or more embodiments of the present disclosure introduces an automated and intelligent data processing method. First, the user feedback data obtained not only includes the current feedback data, but also includes the feedback correlation number, which ensures the diversity and mineability of the user feedback data and lays the foundation for accurate and efficient intelligent data processing. Secondly, the user feedback data is pre-processed and semantically parsed to ensure the readability and accuracy of the parsed data and improve the data processing efficiency. Subsequently, the parsed data is comprehensively analyzed through the problem analysis model, which makes full use of the excellent big data analysis capabilities of the artificial intelligence model and can efficiently generate accurate problem reports. Finally, for the problem report, a problem work order can be automatically generated and distributed to ensure the processing efficiency of user feedback problems.

[0010] The technical solution provided by one or more embodiments of the present disclosure can automatically collect, analyze and process customer feedback issues, thereby improving the efficiency of processing customer feedback issues. By performing semantic parsing on customer feedback data and inputting the parsed data into the problem analysis model, customer feedback issues can be intelligently analyzed, and the real needs of customers can be captured in a timely manner, which is conducive to the customer service terminal to flexibly formulate response plans. The technical solution provided by one or more embodiments of the present disclosure is an excellent SCRM solution that can efficiently and accurately handle customer needs in different scenarios in the online car-hailing industry, and improve customer service quality and user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The features and advantages of the various embodiments of the present disclosure will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the present disclosure in any way. In the accompanying drawings:

[0012] Figure 1 A schematic diagram showing the steps of a method for intelligently processing user feedback in one embodiment of the present disclosure is shown;

[0013] Figure 2 A schematic diagram of the workflow of a user feedback intelligent processing system in one embodiment of the present disclosure is shown;

[0014] Figure 3 A schematic diagram of the workflow of a user feedback intelligent processing system in a practical application scenario of the present disclosure is shown;

[0015] Figure 4 A schematic diagram of functional modules of a user feedback intelligent processing device in one embodiment of the present disclosure is shown;

[0016] Figure 5 A schematic diagram of functional modules of another user feedback intelligent processing device in one embodiment of the present disclosure is shown;

[0017] Figure 6 A schematic structural diagram of an electronic device in one embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0018] To make the purpose, technical solutions, and advantages of the embodiments of the present disclosure more clear, the technical solutions in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments of the present disclosure, all other embodiments obtained by those skilled in the art without making any creative efforts shall fall within the scope of protection of the present disclosure.

[0019] See also Figure 1 The user feedback intelligent processing method provided in one embodiment of the present disclosure can be used in the online car-hailing industry and may include the following steps.

[0020] S1: Obtain user feedback data, where the user feedback data includes current feedback data and feedback-related data, where the feedback-related data includes at least one of user attribute data, user order data, user operation data, historical interaction data, and industry news data.

[0021] In this embodiment, user feedback data can be uploaded by users through the online ride-hailing platform. Through certain application programming interfaces (APIs), the execution subject of this method can automatically collect user feedback data updated in real time on the online ride-hailing platform, ensuring the real-time and completeness of user feedback data. User feedback data can include, but is not limited to, text, images, audio, video, and other media formats.

[0022] In this embodiment, the user feedback data may include not only the current feedback data but also feedback-related data, thereby ensuring the diversity and mineability of the user feedback data and laying the foundation for accurate and efficient intelligent data processing.

[0023] Specifically, current feedback data refers to the feedback data currently provided by users, which can directly reflect their actual needs. For example, text reviews, voice complaints, and photo evidence of the trip. The content of current feedback data may include, but is not limited to, evaluation information (such as a clean interior environment and excellent driver service attitude), consultation information (such as cost deviation consultation and reasons for route changes), and complaint information (such as poor vehicle hygiene and driver violations).

[0024] Feedback-related data can indirectly reflect the user's real needs and improve user satisfaction in the customer service process.

[0025] Feedback-related data can include user attribute data, such as user membership level, historical complaint records, historical consumption frequency and historical consumption amount, etc. Through user attribute data, we can better understand user needs and preferences, thereby providing more targeted customer service.

[0026] Feedback-related data can include user order data, such as trip start and end times, route history, fare details, driver service ratings, etc. User order data can be used to analyze user feedback issues in a targeted manner.

[0027] Feedback-related data can include user operation data, such as the ride-hailing app's operation logs and order operation frequency. This user operation data can reveal the pain points and inconveniences users experience when using the ride-hailing app.

[0028] Feedback-related data can include historical interaction data, such as phone recordings, online chat logs, and complaint handling records. This historical interaction data can help us understand users' past consultation questions and handling results, connect the dots, and optimize customer service response mechanisms and service processes.

[0029] Feedback-related data can include industry news data, such as hot topics related to the ride-hailing industry and hot discussions related to the ride-hailing brands used by users, as reported by various news media. This industry news data can help us understand industry trends, uncover potential user concerns, plan customer service responses in advance, and proactively address user concerns.

[0030] It should be noted that the acquisition of feedback-related data is based on user authorization.

[0031] S2: Perform data preprocessing and semantic analysis on the user feedback data to generate analysis data.

[0032] In this embodiment, the collected user feedback data is preprocessed, which may include data cleaning, format conversion, sentiment analysis preprocessing, etc. By removing noise, filling missing values, and standardizing data formats, the data quality of the analyzed data can be improved, providing a reliable data foundation for subsequent analysis.

[0033] In this embodiment, by introducing some natural language processing technologies and machine learning algorithms, intelligent processing of user feedback data can be achieved, which not only improves the efficiency of generating parsed data but also ensures the accuracy of parsed data.

[0034] In some implementations, a text parsing model can be used to identify intent and analyze user sentiment in text-based user feedback data. This text parsing model can include an entity knowledge graph for the ride-hailing industry. This entity knowledge graph can include entity concepts closely related to the ride-hailing industry, such as "detour," "fare increase," and "rejection of ride."

[0035] In a practical application example, various neural network models, attention mechanism models, and combinations thereof can be used to analyze text data in user feedback. For example, a combination of the BERT model and the BiLSTM model is a preferred text parsing model. The BERT model provides deep, bidirectional contextual information, enabling a better understanding of the semantics of text. The BiLSTM model effectively captures long-term and short-term dependencies in text sequences. By combining BERT's powerful contextual representation with the sequence modeling capabilities of the BiLSTM, text data can be parsed more accurately.

[0036] In some implementations, for the voice type data in the user feedback data, intention recognition can be achieved through automatic speech recognition technology, and sentiment analysis can be performed through voiceprint recognition technology.

[0037] In a practical application example, the voice-type data in user feedback data can be converted through Automatic Speech Recognition (ASR) to easily extract user intent. Simultaneously, voiceprint recognition of the voice-type data can be used to determine the intensity of the user's emotions (e.g., excitement, calmness, etc.). The sentiment analysis results obtained through voiceprint recognition technology can be cross-validated with those obtained from text parsing models to more accurately reflect the user's emotional state.

[0038] In some implementations, an object detection model may be used to identify visual feature data for image type data in user feedback data.

[0039] In a practical application example, the YOLOv5 object detection model can be used for image-type data in user feedback data to identify visual evidence such as vehicle damage and dangerous driving behavior.

[0040] In some embodiments, the data preprocessing and semantic parsing of the user feedback data includes: based on historical feedback data, combined with a time series prediction model, predicting complaint hotspots caused by temporal factors; wherein the temporal factors include at least one of weather factors, holiday factors, and rule change factors.

[0041] In a practical application example, the Prophet model can be combined with the Attention-LSTM model to construct a time series prediction model. This model can be used to predict hot spots for user complaints at specific time points. For example, weather factors such as heavy rain and high temperatures can significantly alter a user's car experience; holidays can significantly alter a user's car experience; and price adjustments, promotions, and other regulatory changes can also significantly alter a user's car experience.

[0042] S3: Analyze the parsed data using the problem analysis model to generate a problem report.

[0043] In this embodiment, a pre-trained question analysis model is used to conduct in-depth analysis of parsed data. The question analysis model can identify key information in the parsed data, such as question type and user sentiment, and generate an analysis report. This report can include both a content analysis of the user's feedback and suggested responses. In particular, the question analysis model can predict potential future user feedback based on historical user feedback, providing proactive decision support for the customer service terminal.

[0044] In some embodiments, the use of the problem analysis model to analyze the parsed data includes: using the problem analysis model to read an industry knowledge base to analyze the parsed data, the industry knowledge base including a basic layer knowledge base, a scenario layer knowledge base, and a strategy layer knowledge base; wherein the basic layer knowledge base is used to provide industry protocol information, the scenario layer knowledge base is used to provide complaint processing logic information, and the strategy layer knowledge base is used to provide differentiated processing strategy information.

[0045] Specifically, the industry knowledge base can be pre-built and can be maintained and updated. The basic layer knowledge base can provide information such as traffic regulations and platform service agreement terms. The scenario layer knowledge base can provide a typical complaint scenario decision tree (for example, if a driver detour is detected, a GPS trajectory comparison decision is triggered). The strategy layer knowledge base can provide a differentiated processing strategy library to meet the regulatory requirements of different cities or regions.

[0046] S4: Generate a problem work order based on the problem report, and send the problem work order to the customer service terminal.

[0047] In this embodiment, the problem report generated by the intelligent analysis model can be automatically classified and labeled, and a corresponding problem ticket can be generated. The problem ticket is automatically sent to the customer service terminal, allowing the corresponding processing personnel to promptly address the user's feedback. The customer service terminal can be either a human or a machine customer service. After receiving the problem ticket, the customer service terminal can use the problem report to understand the cause and effect relationship of the user's feedback problem. The customer service terminal can also provide users with reliable response solutions based on the decision-making suggestions in the problem report.

[0048] In some embodiments, sending the problem work order to the customer service terminal includes: determining the problem type and user emotion corresponding to the problem work order based on the problem report; determining the work order priority based on the problem type and the user emotion; and sending the problem work order to the customer service terminal based on the work order priority.

[0049] Based on factors such as the urgency and importance of user feedback data, the processing priority of problem tickets can be dynamically adjusted to ensure that high-priority user feedback issues are handled in a timely manner, thereby improving user satisfaction.

[0050] In some embodiments, determining the priority of the work order based on the problem type and the user emotion includes: using a multidimensional scoring model to combine the problem type and the user emotion to perform multidimensional scoring on the problem work order; determining the priority of the work order based on the multidimensional scoring results; wherein the scoring dimensions of the multidimensional scoring model include at least one of the personal safety dimension, the economic loss dimension, the transmission risk dimension, and the user value dimension.

[0051] In a practical application example, using a multi-dimensional scoring model, determining the priority of a work order may include the following: in the personal safety dimension, if the problem type involves the user's personal safety (such as a sexual harassment complaint), a red alert will be triggered immediately and transferred to manual review within 10 seconds; in the economic loss dimension, if the problem type involves a single complaint amount that is too large (for example, greater than 500 yuan), or the problem type is a collective loss, it needs to be upgraded; in the communication risk dimension, if the problem type contains public opinion keywords such as "exposure" and "media", it needs to attract extra attention; in the user value dimension, if the user feedback data comes from member users (for example, users with a high annual cumulative consumption ranking), it can be appropriately prioritized. In addition, if the analysis shows that the user's emotions are obviously excessive, these users should be effectively appeased as soon as possible.

[0052] In some embodiments, sending the problem work order to the customer service terminal includes: monitoring the current work order queue and customer service load status; making a work order allocation decision based on the current work order queue, the customer service load status, and the work order priority; sending the problem work order to the customer service terminal in accordance with the work order allocation decision; and optimizing the work order allocation decision based on the processing feedback from the customer service terminal.

[0053] In a practical application example, ticket priorities can be updated every five minutes. Tickets must be re-prioritized immediately if any of the following occurs: the same user submits three or more similar complaints within 15 minutes; industry news data indicates that social media public opinion monitoring indicates a 200% increase in negative topics; or sudden weather changes lead to an abnormal increase in regional orders (e.g., heavy rain at an airport triggering widespread delay complaints).

[0054] In one embodiment of the present disclosure, a method for intelligently processing user feedback is provided. In addition to executing steps S1-S4 above, step S5 may also be executed. Step S5 includes: receiving first satisfaction feedback from the user regarding the problem ticket; receiving second satisfaction feedback from the customer service terminal regarding the problem ticket; and optimizing the problem analysis model based on the first satisfaction feedback and / or the second satisfaction feedback.

[0055] In this embodiment, the client terminal can provide customer service responses to users based on the problem ticket and reference the problem report. User satisfaction data can be collected based on the actual effectiveness of the customer response. This user satisfaction data can be used to optimize the problem analysis model, making problem reports more accurate. Furthermore, the customer service terminal can also evaluate the satisfaction of the problem report and provide suggestions for optimizing the problem analysis model.

[0056] By building a user satisfaction model and quantitatively evaluating customer service processing results, we can provide data support for optimizing the user feedback intelligent processing system. Based on user satisfaction feedback, we can continuously optimize the user feedback intelligent processing system and enhance the user experience.

[0057] In a practical application example, a dual-loop satisfaction feedback mechanism can be established. In the inner loop satisfaction feedback mechanism, customer service processing records can be analyzed daily, and the parameters of the problem analysis model can be updated by comparing the differences between the "system recommendation strategy" and the "manual correction strategy." In the outer loop satisfaction feedback mechanism, cases where user satisfaction drops by more than a certain threshold (for example, 15%) can be retrospectively reviewed every month to correct the problem analysis model.

[0058] See also Figure 2 One embodiment of the present disclosure provides an intelligent user feedback processing system, which may include a data collection module, a preprocessing module, an intelligent analysis module, an intelligent processing module, and a feedback optimization module. The specific functions of these modules correspond to steps S1 through S5 of the intelligent user feedback processing method described above. When applied to the ride-hailing industry, the specific workflow of this intelligent user feedback processing system may be as follows.

[0059] First, customers upload feedback data to the ride-hailing platform in real time. The feedback data can include text, images, voice, etc. The ride-hailing platform can inform users whether the feedback data has been uploaded successfully.

[0060] The data collection module can synchronize data with the online ride-hailing platform, automatically collect customer feedback data on the online ride-hailing platform, and ensure the real-time and completeness of customer feedback data.

[0061] The preprocessing module can preprocess and semantically parse the collected customer feedback data to generate parsed data.

[0062] The intelligent analysis module uses natural language processing, machine learning and other intelligent data processing technologies to conduct in-depth analysis of the parsed customer feedback data and generate problem reports.

[0063] The intelligent processing module automatically creates a problem ticket based on the results of the intelligent analysis module and assigns the problem ticket to the customer service terminal of the online car-hailing platform for subsequent processing.

[0064] The customer service terminal of the online ride-hailing platform provides customer service responses based on the problem tickets assigned by the intelligent processing module and the analysis results of the intelligent analysis module.

[0065] The Feedback Optimization module collects customer satisfaction data on customer service responses and uses this data to adjust and optimize the entire user feedback intelligent processing system. The Feedback Optimization module builds a user satisfaction model and quantitatively evaluates customer satisfaction data, providing data support for system optimization.

[0066] See also Figure 3 ,Facing a practical application scenario, the workflow of the user feedback ,intelligent processing system can be as follows.

[0067] Imagine a ride-hailing user submits a complaint to a ride-hailing platform, stating, "Today's driver was extremely rude. He was late and took a detour. I hope the platform will take this seriously." The platform can then notify the user that the complaint was successful, and the user feedback intelligent processing system will then intelligently handle the complaint.

[0068] The data collection module can collect the complaint information from the online ride-hailing platform through the API interface. The pre-processing module can automatically clean the text of the complaint information and remove meaningless characters. The intelligent analysis module can classify the complaint information as a "service attitude" issue and mark the user's emotion as "negative." Through natural language processing, the intelligent analysis module can extract keywords from the complaint information including "driver's bad attitude", "late" and "detour". The intelligent processing module can automatically generate a complaint work order based on the analysis results of the intelligent analysis module. According to preset rules, the intelligent processing module can automatically distribute the complaint work order to the customer service terminal of the online ride-hailing platform.

[0069] The customer service terminal of the online ride-hailing platform can provide customer service responses to users based on the problem tickets assigned by the intelligent processing module and the analysis results of the intelligent analysis module.

[0070] Finally, the feedback optimization module of the user feedback intelligent processing system can also collect user satisfaction data, thereby optimizing system functions and improving user experience.

[0071] The technical solution provided by one or more embodiments of the present disclosure introduces an automated and intelligent data processing method. First, the user feedback data obtained not only includes the current feedback data, but also includes the feedback correlation number, which ensures the diversity and mineability of the user feedback data and lays the foundation for accurate and efficient intelligent data processing. Secondly, the user feedback data is pre-processed and semantically parsed to ensure the readability and accuracy of the parsed data and improve the data processing efficiency. Subsequently, the parsed data is comprehensively analyzed through the problem analysis model, which makes full use of the excellent big data analysis capabilities of the artificial intelligence model and can efficiently generate accurate problem reports. Finally, for the problem report, a problem work order can be automatically generated and distributed to ensure the processing efficiency of user feedback problems.

[0072] The technical solution provided by one or more embodiments of the present disclosure can automatically collect, analyze and process customer feedback issues, thereby improving the efficiency of processing customer feedback issues. By performing semantic parsing on customer feedback data and inputting the parsed data into the problem analysis model, customer feedback issues can be intelligently analyzed, and the real needs of customers can be captured in a timely manner, which is conducive to the customer service terminal to flexibly formulate response plans. The technical solution provided by one or more embodiments of the present disclosure is an excellent SCRM solution that can efficiently and accurately handle customer needs in different scenarios in the online car-hailing industry, and improve customer service quality and user satisfaction.

[0073] See also Figure 4 The present disclosure further provides a user feedback intelligent processing device, the device comprising:

[0074] The data acquisition unit 100 is configured to acquire user feedback data, wherein the user feedback data includes current feedback data and feedback-related data, wherein the feedback-related data includes at least one of user attribute data, user order data, user operation data, historical interaction data, and industry news data;

[0075] The data parsing unit 200 is used to perform data preprocessing and semantic parsing on the user feedback data to generate parsed data;

[0076] The problem analysis unit 300 is used to analyze the parsed data using a problem analysis model and generate a problem report;

[0077] The work order processing unit 400 is used to generate a problem work order according to the problem report and send the problem work order to the customer service terminal.

[0078] In one embodiment, the data parsing unit 200 includes a first parsing subunit 201. The first parsing subunit 201 is specifically configured to: perform intent recognition and sentiment analysis on text data in the user feedback data using a text parsing model, wherein the text parsing model includes an entity knowledge graph of the ride-hailing industry; perform intent recognition on voice data in the user feedback data using automatic speech recognition technology and perform sentiment analysis using voiceprint recognition technology; and perform visual feature recognition on image data in the user feedback data using an object detection model.

[0079] In one embodiment, the data analysis unit 200 includes a second analysis subunit 202. The second analysis subunit 202 is specifically configured to predict complaint hotspots caused by temporal factors based on historical feedback data and in combination with a time series prediction model; wherein the temporal factors include at least one of weather factors, holiday factors, and rule change factors.

[0080] In one embodiment, the problem analysis unit 300 is specifically used to use the problem analysis model to read the industry knowledge base and analyze the parsed data. The industry knowledge base includes a basic layer knowledge base, a scenario layer knowledge base and a strategy layer knowledge base; wherein the basic layer knowledge base is used to provide industry protocol information, the scenario layer knowledge base is used to provide complaint processing logic information, and the strategy layer knowledge base is used to provide differentiated processing strategy information.

[0081] In one embodiment, the work order processing unit 400 is specifically used to determine the problem type and user emotion corresponding to the problem work order based on the problem report; determine the work order priority based on the problem type and the user emotion; and send the problem work order to the customer service terminal based on the work order priority.

[0082] In one embodiment, the work order processing unit 400 includes a priority calculation subunit 401. The priority calculation subunit 401 is specifically configured to use a multidimensional scoring model, combined with the problem type and the user's emotions, to perform a multidimensional scoring on the problem work order; and determine the work order priority based on the multidimensional scoring results; wherein the scoring dimensions of the multidimensional scoring model include at least one of a personal safety dimension, an economic loss dimension, a transmission risk dimension, and a user value dimension.

[0083] See also Figure 5 In one embodiment, the user feedback intelligent processing device further includes a feedback optimization unit 500. The feedback optimization unit 500 is specifically configured to receive a first satisfaction feedback from a user regarding the problem ticket; receive a second satisfaction feedback from a customer service terminal regarding the problem ticket; and optimize the problem analysis model based on the first satisfaction feedback and / or the second satisfaction feedback.

[0084] The various units described in the above embodiments can be implemented by computer chips or products with certain functions. A typical implementation device is a computer. Specifically, the computer can be, for example, a personal computer, a laptop computer, a cellular phone, a camera phone, a smartphone, a personal digital assistant, a media player, a navigation device, an email device, a game console, a tablet computer, a wearable device, or a combination of any of these devices.

[0085] For the convenience of description, the above devices are described as being divided into various units according to their functions. Of course, when implementing this application, the functions of each unit can be implemented in the same or multiple software and / or hardware.

[0086] See also Figure 6The present disclosure also provides an electronic device, which includes a memory and a processor, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, the above-mentioned user feedback intelligent processing method is implemented.

[0087] The present disclosure also provides a computer-readable storage medium, which is used to store a computer program. When the computer program is executed by a processor, it implements the above-mentioned user feedback intelligent processing method.

[0088] The processor may be a central processing unit (CPU). The processor may also be other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, or a combination of the above chips.

[0089] Memory, as a non-transitory computer-readable storage medium, can be used to store non-transitory software programs, non-transitory computer executable programs, and modules, such as the program instructions / modules corresponding to the methods in the embodiments of the present disclosure. The processor executes the non-transitory software programs, instructions, and modules stored in the memory to perform various processor functions and data processing, thereby implementing the methods in the aforementioned method embodiments.

[0090] The memory may include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function; the data storage area may store data created by the processor, etc. In addition, the memory may include a high-speed random access memory, and may also include a non-transitory memory, such as at least one disk storage device, a flash memory device, or other non-transitory solid-state storage device. In some embodiments, the memory may optionally include a memory remotely located relative to the processor, and these remote memories may be connected to the processor via a network. Examples of the above-mentioned network include, but are not limited to, the Internet, an intranet, a local area network, a mobile communication network, and combinations thereof.

[0091] Those skilled in the art will appreciate that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory, a hard disk drive (HDD), or a solid-state drive (SSD). The storage medium can also include a combination of the above-mentioned types of memory.

[0092] The various embodiments in this specification are described in a progressive manner. Similar portions between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from the other embodiments. In particular, the device, equipment, and storage medium embodiments are generally similar to the method embodiments, so their descriptions are relatively simplified. For relevant portions, refer to the descriptions of the method embodiments.

[0093] The foregoing is merely an embodiment of the present application and is not intended to limit the present application. For those skilled in the art, the present application may have various changes and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application should all be included within the scope of the claims of the present application.

[0094] Although the embodiments of the present disclosure have been described with reference to the accompanying drawings, those skilled in the art may make various modifications and variations without departing from the spirit and scope of the present disclosure, and such modifications and variations are all within the scope defined by the appended claims.

Claims

1. A method for intelligently processing user feedback, characterized in that: The method is used in the online car-hailing industry and includes: Acquire user feedback data, where the user feedback data includes current feedback data and feedback-related data, where the feedback-related data includes at least one of user attribute data, user order data, user operation data, historical interaction data, and industry news data; Performing data preprocessing and semantic analysis on the user feedback data to generate analysis data; Analyzing the parsed data using a problem analysis model to generate a problem report; A problem work order is generated based on the problem report, and the problem work order is sent to a customer service terminal.

2. The method according to claim 1, characterized in that The performing data preprocessing and semantic parsing on the user feedback data includes at least one of the following: For the text type data in the user feedback data, a text parsing model is used to perform intent recognition and sentiment analysis, wherein the text parsing model includes an entity knowledge graph of the online ride-hailing industry; For the voice type data in the user feedback data, automatic speech recognition technology is used to realize intention recognition, and voiceprint recognition technology is used to perform sentiment analysis; For the image type data in the user feedback data, an object detection model is used to identify visual feature data.

3. The method according to claim 1, characterized in that The performing data preprocessing and semantic parsing on the user feedback data includes: Based on historical feedback data and combined with time series prediction models, we can predict complaint hotspots caused by temporal factors. The timing factors include at least one of weather factors, holiday factors and rule change factors.

4. The method according to claim 1, wherein The analyzing the parsed data using the problem analysis model includes: Utilizing the problem analysis model, reading the industry knowledge base and analyzing the parsed data, the industry knowledge base includes a basic layer knowledge base, a scenario layer knowledge base, and a strategy layer knowledge base; Among them, the basic layer knowledge base is used to provide industry protocol information, the scenario layer knowledge base is used to provide complaint processing logic information, and the strategy layer knowledge base is used to provide differentiated processing strategy information.

5. The method according to claim 1, wherein The sending of the problem work order to the customer service terminal includes: Determine the problem type and user sentiment corresponding to the problem ticket based on the problem report; Determine the priority of the work order based on the problem type and the user sentiment; According to the priority of the work order, the problem work order is sent to the customer service terminal.

6. The method according to claim 5, characterized in that Determining the work order priority based on the problem type and the user sentiment includes: Using a multidimensional scoring model, combined with the problem type and the user sentiment, to perform a multidimensional scoring on the problem ticket; Determine the priority of the work order based on the multi-dimensional scoring results; Among them, the scoring dimensions of the multidimensional scoring model include at least one of the personal safety dimension, economic loss dimension, communication risk dimension, and user value dimension.

7. The method according to claim 1, characterized in that The method further comprises: Receive first satisfaction feedback from the user regarding the problem ticket; Receiving second satisfaction feedback from the customer service terminal regarding the problem work order; Optimize the problem analysis model based on the first satisfaction feedback and / or the second satisfaction feedback.

8. A user feedback intelligent processing device, characterized in that: The device is used in the online car-hailing industry and includes: a data acquisition unit, configured to acquire user feedback data, wherein the user feedback data includes current feedback data and feedback-related data, wherein the feedback-related data includes at least one of user attribute data, user order data, user operation data, historical interaction data, and industry news data; A data parsing unit, configured to perform data preprocessing and semantic parsing on the user feedback data to generate parsed data; A problem analysis unit, configured to analyze the parsed data using a problem analysis model and generate a problem report; The work order processing unit is used to generate a problem work order based on the problem report and send the problem work order to the customer service terminal.

9. An electronic device, characterized in that: The electronic device includes a memory and a processor, the memory is used to store a computer program, and when the computer program is executed by the processor, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images