Customer complaint information processing method and device, equipment and storage medium

Abstract

The invention provides a customer complaint information processing method and device, equipment and a storage medium. The method comprises the steps that customer complaint information sent by a user side is received, text information in the customer complaint information is preprocessed to obtain text information in a preset format, feature extraction is conducted on user information in the customer complaint information to obtain corresponding user feature information, the user feature information is input into an emotion analysis model, and an emotion analysis result is obtained; according to the user emotion tendency, outputting the corresponding user emotion tendency, determining the processing priority of the customer complaint information based on the user emotion tendency, and adding the customer complaint information into a to-be-processed queue; inputting text information in the customer complaint information in the to-be-processed queue into the semantic analysis model in sequence according to the processing priorities at preset specified time, so as to output corresponding solutions; and sending the obtained solution to the user side, so that the user side displays the solution. Based on the method provided by the invention, the flexibility of processing the customer complaint information can be improved.

Description

Technical Field

[0001] The present application relates to the field of data processing technology, and in particular to a method, device, equipment and storage medium for processing customer complaint information. Background Art

[0002] With the development of financial technology, banks and financial institutions are increasingly relying on technology to improve customer service and business processes. The resulting problem is that the number of customer complaints received has increased and the types of customer complaints are diverse. Most of the customer complaint information received is in the form of text or voice.

[0003] At present, in the prior art, for customer complaint information in text form, the customer complaint information is processed in order of the time of the customer complaint to provide a corresponding solution.

[0004] However, the existing method of processing customer complaint information in text form is to mechanically process the customer complaint information in the order of the time of the customer complaint, and lacks flexibility in adjusting the customer complaint information that needs to be processed urgently, resulting in untimely processing of some urgent customer complaint information. Summary of the invention

[0005] The present application provides a customer complaint information processing method, device, equipment and storage medium, which are used to solve the problem of untimely processing of urgent customer complaint information in the prior art and improve the flexibility of processing customer complaint information.

[0006] In a first aspect, the present application provides a customer complaint information processing method, which is applied to a server, comprising: receiving customer complaint information sent by a user terminal, wherein the customer complaint information is generated by the user terminal in response to a customer complaint operation of the user, wherein the customer complaint information includes user information and text information;

[0007] Preprocessing the text information in the customer complaint information to obtain text information in a preset format;

[0008] Extracting features from the user information in the customer complaint information to obtain corresponding user feature information;

[0009] Inputting the text information in a preset format in the customer complaint information and the corresponding user feature information into a sentiment analysis model to output the corresponding user sentiment tendency;

[0010] Determining a processing priority of the customer complaint information based on the user sentiment tendency corresponding to the customer complaint information;

[0011] Add the customer complaint information to a queue to be processed;

[0012] At a preset designated time, the text information in the customer complaint information in the queue to be processed is input into the semantic analysis model in order according to the processing priority, so as to output the corresponding solution;

[0013] The obtained solution is sent to the user end so that the user end displays the solution.

[0014] In a possible design, the text information in a preset format in the customer complaint information and the corresponding user feature information are input into a sentiment analysis model to output the corresponding user sentiment tendency, including:

[0015] Inputting the text information in a preset format in the customer complaint information and the corresponding user feature information into a sentiment analysis model, so that the sentiment analysis model performs feature extraction on the text information to output a first feature vector corresponding to the text information, and performs feature extraction on the user feature information to output a second feature vector corresponding to the user feature information;

[0016] Determine a first weight corresponding to the text information and a second weight corresponding to the user characteristic information;

[0017] Adding the product of the first feature vector and the first weight to the product of the second feature vector and the first weight to obtain a fused feature vector;

[0018] The fused feature vector is input into the sentiment analysis model to output the corresponding sentiment tendency.

[0019] In a possible design, determining the processing priority of the customer complaint information based on the user emotional tendency corresponding to the customer complaint information includes:

[0020] Identifying the type of the user's emotional tendency, wherein the type includes positive tendency, negative tendency and neutral tendency;

[0021] If the user's emotional tendency is positive, determining the processing priority of the customer complaint information to be the first processing priority;

[0022] If the user's emotional tendency is neutral, determining the processing priority of the customer complaint information to be the second processing priority;

[0023] If the user's emotional tendency is negative, the processing priority of the customer complaint information is determined to be a third processing priority, wherein the third processing priority is higher than the second processing priority, and the second processing priority is higher than the first processing priority.

[0024] In a possible design, the customer complaint information includes the customer complaint time;

[0025] Accordingly, after determining the processing priority of the customer complaint information based on the user emotional tendency corresponding to the customer complaint information, the method further includes:

[0026] If there are two or more customer complaints with the same processing priority, the customer complaint information will be input into the semantic analysis model in order from the earliest to the latest time of the customer complaint to output the corresponding solution.

[0027] In a possible design, the training process of the sentiment analysis model includes:

[0028] Acquiring first training data, wherein the first training data includes historical customer complaint information and standard sentiment orientation;

[0029] Inputting the historical customer complaint information into a sentiment analysis model so that the sentiment analysis model outputs a predicted sentiment tendency;

[0030] Based on the loss function, the predicted sentiment tendency is compared with the standard sentiment tendency to obtain the corresponding loss value;

[0031] If the loss value exceeds a preset threshold, adjusting the parameters of the sentiment analysis model;

[0032] The steps from inputting the historical customer complaint information into the sentiment analysis model to adjusting the parameters of the sentiment analysis model are executed in a loop until the loss value between the predicted sentiment tendency output by the sentiment analysis model and the standard sentiment tendency does not exceed the preset threshold, or the loop is stopped when the number of times reaches the preset threshold, so as to obtain a trained sentiment analysis model.

[0033] In a possible design, the training process of the semantic analysis model includes:

[0034] Acquiring second training data, wherein the second training data includes historical customer complaint information and standard solutions;

[0035] Inputting the historical customer complaint information into a semantic analysis model so that the semantic analysis model outputs a prediction solution;

[0036] Compare the predicted solution with the standard solution based on the loss function to obtain the corresponding loss value;

[0037] If the loss value exceeds a preset threshold, adjusting the parameters of the semantic analysis model based on a back propagation algorithm;

[0038] The steps from inputting the historical customer complaint information into the semantic analysis model to adjusting the parameters of the semantic analysis model based on the back propagation algorithm are executed cyclically until the loss value between the predicted solution output by the semantic analysis model and the standard solution does not exceed the preset threshold, or the cycle is stopped when the number of cycles reaches the preset threshold, so as to obtain a trained semantic analysis model.

[0039] In one possible design, it also includes:

[0040] Receiving feedback information sent by a user terminal, wherein the feedback information is generated by the user terminal in response to a user's feedback operation on a solution;

[0041] Count the values ​​of various indicators in the feedback information within a preset time period;

[0042] If the value of the indicator is lower than the preset value, the sentiment analysis model and the semantic analysis model will be optimized and trained.

[0043] In a second aspect, the present application provides a customer complaint information processing device, which is applied to a server, including:

[0044] A receiving module, configured to receive customer complaint information sent by a user terminal, wherein the customer complaint information is generated by the user terminal in response to a customer complaint operation of the user, and wherein the customer complaint information includes user information and text information;

[0045] A processing module, used for preprocessing the text information in the customer complaint information to obtain text information in a preset format;

[0046] The processing module is further used to extract features from the user information in the customer complaint information to obtain corresponding user feature information;

[0047] An input module, used to input the text information in a preset format in the customer complaint information and the corresponding user feature information into the sentiment analysis model to output the corresponding user sentiment tendency;

[0048] The processing module is further used to determine the processing priority of the customer complaint information based on the user emotional tendency corresponding to the customer complaint information;

[0049] The processing module is also used to add the customer complaint information to a queue to be processed;

[0050] The input module is further used to input the text information in the customer complaint information in the queue to be processed into the semantic analysis model in sequence according to the processing priority at a preset specified time, so as to output a corresponding solution;

[0051] The processing module is further used to send the obtained solution to the user terminal so that the user terminal displays the solution.

[0052] In a third aspect, an embodiment of the present application provides an electronic device, comprising: at least one processor and a memory; the memory stores computer-executable instructions; the at least one processor executes the computer-executable instructions stored in the memory, so that the at least one processor executes the customer complaint information processing method described in the first aspect and various possible designs of the first aspect.

[0053] In a fourth aspect, an embodiment of the present application provides a computer-readable storage medium, in which computer execution instructions are stored. When a processor executes the computer execution instructions, the customer complaint information processing method described in the first aspect and various possible designs of the first aspect is implemented.

[0054] In a fifth aspect, an embodiment of the present application provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the customer complaint information processing method described in the first aspect and various possible designs of the first aspect.

[0055] The customer complaint information processing method, device, equipment and storage medium provided by the present application first receive the customer complaint information sent by the user end, pre-process the text information in the customer complaint information into text information in a preset format, extract the features of the user information to obtain the corresponding user feature information, then input the text information in the preset format and the corresponding user feature information into the sentiment analysis model to obtain the corresponding user sentiment tendency, determine the processing priority of the customer complaint information by the user sentiment tendency, add the customer complaint information to the queue to be processed, and at a preset specified time, input the corresponding text information of all the customer complaint information in the queue to be processed into the semantic analysis model in order according to the processing priority to obtain the corresponding solution, and finally send the obtained solution to the user end so that the user end displays the solution. Therefore, based on the method provided by the present application, customer complaint information with a high processing priority can be processed first to improve the flexibility of processing customer complaint information. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0057] Figure 1 This is the application scenario diagram of this application;

[0058] Figure 2 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 1 ;

[0059] Figure 3 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 2 ;

[0060] Figure 4 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 3 ;

[0061] Figure 5 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 4 ;

[0062] Figure 6 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 5 ;

[0063] Figure 7 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 6 ;

[0064] Figure 8 A schematic diagram of the structure of a customer complaint information processing device provided in an embodiment of the present application;

[0065] Fig. 9 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application.

[0066] The above drawings have shown clear embodiments of the present application, which will be described in more detail later. These drawings and text descriptions are not intended to limit the scope of the present application in any way, but to illustrate the concept of the present application to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0067] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The implementations described in the following exemplary embodiments do not represent all implementations consistent with the present application. Instead, they are merely examples of devices and methods consistent with some aspects of the present application as detailed in the appended claims.

[0068] In the technical solution of this application, the collection, storage, use, processing, transmission, provision and disclosure of information such as financial data or user data involved shall comply with the provisions of relevant laws and regulations and shall not violate public order and good morals.

[0069] It should be noted that in the embodiments of the present application, certain software, components, models and other existing solutions in the industry may be mentioned, and they should be regarded as exemplary. Their purpose is only to illustrate the feasibility of implementing the technical solution of the present application, but it does not mean that the applicant has or will necessarily use the solution.

[0070] Figure 1 This is the application scenario diagram of this application. Figure 1 As shown, the application scenario includes: the application scenario includes: a user end and a server end. The user end may be an electronic device such as a mobile phone, a tablet, or a computer. The user end is used to send customer complaint information to the server end. The customer complaint information may be the user's demand information. The server end may be a server, which is used to push the solution corresponding to the customer complaint information to the server end.

[0071] In the prior art, for customer complaint information in text form, the customer complaint information is processed in order of the time of the customer complaint to provide corresponding solutions. However, the method of processing customer complaint information in text form in the prior art is just to mechanically process the customer complaint information in the order of the time of the customer complaint, and lacks flexibility in adjusting the customer complaint information that needs to be processed urgently, resulting in untimely processing of some urgent customer complaint information.

[0072] In response to the above technical problems, the present application proposes the following technical concept: first, receive the customer complaint information sent by the user end, pre-process the text information in the customer complaint information into text information in a preset format, extract the features of the user information to obtain the corresponding user feature information, then input the text information in the preset format and the corresponding user feature information into the sentiment analysis model to obtain the corresponding user sentiment tendency, determine the processing priority of the customer complaint information by the user sentiment tendency, add the customer complaint information to the queue to be processed, and at a preset specified time, input the corresponding text information of all the customer complaint information in the queue to be processed into the semantic analysis model in order according to the processing priority to obtain the corresponding solution, and finally send the obtained solution to the user end so that the user end displays the solution. Therefore, based on the method provided by the present application, customer complaint information with a high processing priority can be processed first to improve the flexibility of processing customer complaint information.

[0073] The technical solution of the present application and how the technical solution of the present application solves the above-mentioned technical problems are described in detail below with specific embodiments. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0074] based on Figure 1 In the application scenario shown, an embodiment of the present application also provides a method for processing customer complaint information. Figure 2 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 1 .like Figure 2 As shown, the customer complaint information processing method includes:

[0075] S201. Receive customer complaint information sent by a user terminal, wherein the customer complaint information is generated by the user terminal in response to a customer complaint operation of a user, and wherein the customer complaint information includes user information and text information.

[0076] The execution subject of the method of this embodiment may be Figure 1 The server shown in .

[0077] Combined with the scenario example, the customer complaint information can be the user's demand information, where the text information can be obtained by the user input on the client, and the client can be the user's smart phone or tablet and other electronic products. The client used by the user is bound to the user information, and the user information can be stored locally on the client. Therefore, before the client transmits the text information output by the user to the server, it can send the text information and the locally stored user information as the user's customer complaint information to the server. Therefore, the customer complaint information received by the server includes the text information input by the user on the user terminal and the user information stored locally on the user terminal.

[0078] S202: pre-process the text information in the customer complaint information to obtain text information in a preset format.

[0079] Combined with the scenario example, preprocessing includes data cleaning, missing value processing and standardization. Specifically, the text information in the customer complaint information can be segmented, stop words can be removed, and stemming can be performed, and the processed text information can be converted into a preset format. The preset format represents the format that can be processed by the machine learning algorithm. The machine learning algorithm can process a variety of data formats including but not limited to Comma-Separated Values ​​(CSV), TXT format, PKL format, JavaScript Object Notation (JSON) format, etc.

[0080] S203: Extract features of the user information in the customer complaint information to obtain corresponding user feature information.

[0081] Combined with the scenario examples, user information includes but is not limited to the customer's basic information and financial behavior information, transaction habits, preferences, etc. After feature extraction of user information, the obtained user feature information includes but is not limited to the user's age, occupation, transaction behavior, assets and liabilities, contact information, etc.

[0082] S204: Input the text information in a preset format in the customer complaint information and the corresponding user feature information into a sentiment analysis model to output the corresponding user sentiment tendency.

[0083] Combined with the scenario example, the sentiment analysis model can be a pre-trained large-scale language model or a sentiment classification model customized according to task requirements. The model can combine customer feature information to identify the sentiment tendency corresponding to the text information, that is, the user's sentiment tendency.

[0084] S205: Determine a processing priority of the customer complaint information based on the user emotional tendency corresponding to the customer complaint information.

[0085] Combined with the scenario example, the urgency of the user's customer complaint information is determined according to the user's emotional tendency, and the processing priority of the customer complaint information is determined in turn. Specifically, the more urgent customer complaint information can be processed first, so the processing priority of the more urgent customer complaint information can be increased.

[0086] S206: Add the customer complaint information to a queue to be processed.

[0087] In combination with the scenario example, all unprocessed customer complaints are arranged in the queue to be processed in order. The customer complaints in the queue to be processed are arranged according to the processing priority. The customer complaints with higher processing priority are arranged at the front.

[0088] S207: At a preset designated time, the text information in the customer complaint information in the queue to be processed is input into the semantic analysis model in order according to the processing priority, so as to output a corresponding solution.

[0089] Combined with the scenario example, the preset designated time can be a time period. For example, the designated time can be preset from 9:00 a.m. to 12:00 p.m. every day. During this time period, the customer complaint information in the queue to be processed is input into the semantic analysis model in sequence according to the order of arrangement between the customer complaint information. The semantic analysis model can be selected as BERT, GPT, etc. The semantic analysis model can analyze the customer's demands by analyzing the text information in the customer complaint information, and give corresponding solutions based on the customer's demands, and output the corresponding solutions.

[0090] S208: Send the obtained solution to the user terminal, so that the user terminal displays the solution.

[0091] Combined with the scenario example, the server obtains the solution output by the semantic analysis model and sends the solution to the user end, which can display the obtained solution on the display screen.

[0092] Based on the method provided in the embodiment, customer complaint information with a high priority can be processed preferentially to improve the flexibility of processing customer complaint information.

[0093] Optional, Figure 3 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 2 , S204 includes:

[0094] S301. Input text information in a preset format in the customer complaint information and corresponding user feature information into a sentiment analysis model, so that the sentiment analysis model performs feature extraction on the text information to output a first feature vector corresponding to the text information, and performs feature extraction on the user feature information to output a second feature vector corresponding to the user feature information.

[0095] In combination with the scenario example, the first feature vector represents the feature information of the text information in the customer complaint information, and the first feature vector can be assumed to be X1. The second feature vector represents the user feature information in the customer information, and the second feature vector can be assumed to be X2.

[0096] S302: Determine a first weight corresponding to the text information and a second weight corresponding to the user characteristic information.

[0097] In combination with the scenario example, specific values ​​of the first weight and the second weight can be determined according to actual conditions. For example, the value of the first weight is determined to be a1, and the value of the second weight is determined to be a2.

[0098] S303: Add the product of the first feature vector and the first weight to the product of the second feature vector and the first weight to obtain a fused feature vector.

[0099] Combined with the scenario example, the first eigenvector and the second eigenvector are weightedly summed based on the first weight and the second weight to obtain the sum result. Specifically, the product of the first weight and the first eigenvector is calculated, and the product can be represented as: a1*X1. The product of the second weight and the second eigenvector is calculated, and the product can be represented as: a2*X2. The sum result can be represented as: a1*X1+a2*X2. The final sum result is used as the fused feature vector.

[0100] S304: Input the fused feature vector into the sentiment analysis model to output the corresponding sentiment tendency.

[0101] Combined with the scene examples, the sentiment analysis model analyzes the corresponding sentiment tendency based on the final fused feature vector.

[0102] This example can improve the accuracy of the analysis results of the sentiment analysis model by setting the weights corresponding to the feature vectors of the text information and the weights corresponding to the user feature information.

[0103] Optional, Figure 4 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 3 ,like Figure 4 As shown, S205 includes:

[0104] S401. Identify the type of the user's emotional tendency, where the type includes positive tendency, negative tendency and neutral tendency.

[0105] Combined with the scenario example, the classification of emotional tendency can be divided into three types according to the degree of urgency. The degree of urgency can be based on the time limit information in the text information. The customer complaint information with urgent time limit information has a higher degree of urgency. In this case, the customer complaint information can have a negative tendency. For example, the time limit information can be characterized by a deadline. If the deadline is less than 2 days, the customer complaint information can be determined as a customer complaint information with a higher degree of urgency, and the emotional tendency of the customer complaint information can be determined as a negative tendency. The urgency of the customer complaint information with a relatively ample time limit is lower, and the customer complaint information can be determined as a positive tendency. For example, if the deadline is greater than 5 days, the customer complaint information can be determined as a customer complaint information with a lower degree of urgency, and the emotional tendency of the customer complaint information can be determined as a positive tendency. The emotional tendency between the positive tendency and the negative tendency is determined as a neutral tendency. For example, if the deadline is greater than 2 days and less than 5 days, the emotional tendency of the customer complaint information can be determined as a neutral tendency. If the text information does not include time limit information, the emotional tendency of customer complaint information with a more impatient tone can be determined as negative, the emotional tendency of customer complaint information with a more gentle tone can be determined as positive, and the emotional tendency of customer complaint information with a normal tone can be determined as neutral. Alternatively, the user's personality can be determined by analyzing the user's characteristic information, and the emotional tendency of the customer complaint information can be determined based on the user's personality.

[0106] S402: If the user's emotional tendency is positive, determining the processing priority of the customer complaint information to be the first processing priority.

[0107] Combined with the scenario example, since the positive customer complaint information indicates that it is not urgent, the processing order of the customer complaint information can be relatively low, and the customer complaint information can be determined as the first processing priority, which is the lowest level of processing priority.

[0108] S403: If the user's emotional tendency is neutral, determining the processing priority of the customer complaint information to be the second processing priority.

[0109] Combined with the scenario example, since customer complaint information with a neutral tendency represents customer complaint information that is between non-urgent and urgent, the processing order of this customer complaint information can be ranked before customer complaint information with a positive tendency and after customer complaint information with a negative tendency. This customer complaint information can be determined as the second processing priority, which is higher than the first processing priority.

[0110] 404. If the user's emotional tendency is negative, determine that the processing priority of the customer complaint information is a third processing priority, wherein the third processing priority is higher than the second processing priority, and the second processing priority is higher than the first processing priority.

[0111] In combination with the scenario example, since customer complaint information with a negative tendency represents urgent customer complaint information, the processing order of the customer complaint information can be ranked first, and the customer complaint information can be determined as the third processing priority, which is higher than the second processing priority.

[0112] Based on the method provided in this example, the processing priority of customer complaint information can be determined according to the emotional tendency of the customer complaint information.

[0113] Optionally, the customer complaint information includes the customer complaint time;

[0114] Accordingly, after S205, the following steps are also included:

[0115] If there are two or more customer complaints with the same processing priority, the customer complaint information will be input into the semantic analysis model in order from the earliest to the latest time of the customer complaint to output the corresponding solution.

[0116] Combined with the scenario example, in the queue to be processed, if there are multiple customer complaints with the same processing priority, for example, there are 10 customer complaints with the third processing priority in the queue to be processed, since the urgency of these 10 customer complaints with the third processing priority is the same, the order of processing can be processed in sequence according to the order of customer complaint time. The customer complaint time represents the time when the user end sends the customer complaint information to the server end. The earlier the customer complaint time, the higher the processing order of the customer complaint information.

[0117] Optional, Figure 5 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 4 ,like Figure 5 As shown, the training process of the sentiment analysis model includes:

[0118] S501. Acquire first training data, wherein the first training data includes historical customer complaint information and standard sentiment tendency.

[0119] In combination with the scenario example, the first training data is the training data of the sentiment analysis model, which can be obtained by collecting the user's historical customer complaint information and the corresponding standard sentiment tendency.

[0120] S502: Input the historical customer complaint information into a sentiment analysis model so that the sentiment analysis model outputs a predicted sentiment tendency.

[0121] Combined with the scenario example, the parameters of the sentiment analysis model before training are initialized parameters. After the historical customer complaint information in the first training data is input into the sentiment analysis model, the sentiment analysis model analyzes based on the initialized parameters to give a predicted sentiment tendency.

[0122] S503: Based on the loss function, the predicted emotional tendency is compared with the standard emotional tendency to obtain a corresponding loss value.

[0123] Combined with the scene example, the difference between the predicted emotional tendency obtained by the sentiment analysis model based on the initialized parameter analysis and the standard emotional tendency, that is, the loss value, can be obtained through the loss function.

[0124] S504: If the loss value exceeds a preset threshold, adjust the parameters of the sentiment analysis model.

[0125] Combined with the scenario example, the preset threshold can be a maximum threshold of the loss value preset in advance. If the loss value exceeds the maximum threshold, it means that the prediction deviation of the sentiment analysis model is too large. The initialization parameters of the sentiment analysis model are adaptively adjusted to obtain new parameters, such as optimizing the model parameters through the back propagation algorithm.

[0126] S505. Loop through the steps from inputting the historical customer complaint information into the sentiment analysis model to adjusting the parameters of the sentiment analysis model until the loss value between the predicted sentiment tendency output by the sentiment analysis model and the standard sentiment tendency does not exceed the preset threshold, or stop when the number of loops reaches the preset threshold, so as to obtain a trained sentiment analysis model.

[0127] In combination with the scenario example, the historical customer complaint information in the first training data is re-input into the sentiment analysis model after the parameters are adjusted, so that the sentiment analysis model outputs the predicted emotional image again, and when the loss value exceeds the maximum threshold, the model parameters of the sentiment analysis model are adjusted again. Repeat the above steps until the loss value does not exceed the maximum threshold, then stop training the sentiment analysis model, and use the model parameters obtained by the last adjustment as the parameters of the final sentiment analysis model to obtain a trained sentiment analysis model. Or when the loop reaches a preset threshold, for example, the preset threshold of the number of loops is 1000 times, then after the number of loops reaches 1000, stop training the sentiment analysis model, even if the loss value exceeds the maximum threshold, use the model parameters obtained by the last adjustment as the parameters of the final sentiment analysis model to obtain a trained sentiment analysis model. Based on the method provided in this example, complete the training of the sentiment analysis model to obtain a trained sentiment analysis model.

[0128] Optional, Figure 6 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 5 ,like Figure 6 As shown, the training process of the semantic analysis model includes:

[0129] S601. Acquire second training data, wherein the second training data includes historical customer complaint information and standard solutions.

[0130] Combined with the scenario example, a semantic analysis model is established, including: word embedding, which refers to the process of mapping words to real vector space, converting words into dense vector representations, so that the semantic relationship between words can be reflected in the vector space. In the code, technologies such as Word2Vec and GloVe can be used for word embedding to represent words as vectors of fixed dimensions. The recurrent neural network of the semantic analysis model can be RNN, which can process sequence data. The model type of the semantic analysis model can be Transformer, which is a deep learning model based on the self-attention mechanism, suitable for processing long text sequences and large-scale data. In the code, the Transformer model can be used to build a semantic analysis model to improve the performance of the model on complex text data. The attention mechanism of the semantic analysis model allows the model to pay more attention to the parts related to the current task when processing sequence data, improving the performance of the model on long sequence data. In the code, the attention mechanism can be introduced into models such as RNN or Transformer to enhance the model's ability to understand the input sequence.

[0131] The second training data is the training data of the semantic analysis model, which can be obtained by collecting historical customer complaint information of users and corresponding standard solutions.

[0132] S602: Input the historical customer complaint information into a semantic analysis model, so that the semantic analysis model outputs a prediction solution.

[0133] Combined with the scenario example, the parameters of the semantic analysis model before training are initialized parameters. After the historical customer complaint information in the second training data is input into the semantic analysis model, the semantic analysis model analyzes based on the initialized parameters and gives a prediction solution.

[0134] S603: Compare the predicted solution with the standard solution based on the loss function to obtain a corresponding loss value.

[0135] Combined with the scenario example, the difference between the predicted solution obtained by the semantic analysis model based on the initialized parameter analysis and the standard solution, that is, the loss value, can be obtained through the loss function. Commonly used loss functions can be the cross entropy loss function, the mean square error loss function, etc.

[0136] S604: If the loss value exceeds a preset threshold, the parameters of the semantic analysis model are adjusted based on a back propagation algorithm.

[0137] Combined with the scenario example, the preset threshold can be the maximum threshold of the loss value preset in advance. If the loss value exceeds the maximum threshold, it means that the prediction deviation of the semantic analysis model is too large. The initialization parameters of the semantic analysis model are adaptively adjusted to obtain new parameters, such as optimizing the model parameters through the back propagation algorithm.

[0138] S605. Loop through the steps from inputting the historical customer complaint information into the semantic analysis model to adjusting the parameters of the semantic analysis model based on the back propagation algorithm, until the loss value between the predicted solution output by the semantic analysis model and the standard solution does not exceed the preset threshold, or stop when the number of loops reaches the preset threshold, to obtain a trained semantic analysis model.

[0139] In combination with the scenario example, the historical customer complaint information in the second training data is re-input into the semantic analysis model after parameter adjustment, so that the semantic analysis model outputs a prediction solution again, and when the loss value exceeds the maximum threshold, the model parameters of the semantic analysis model are adjusted again. Repeat the above steps until the loss value does not exceed the maximum threshold, then stop training the semantic analysis model, and use the model parameters obtained by the last adjustment as the parameters of the final semantic analysis model to obtain a trained semantic analysis model. Or when the loop reaches a preset threshold, such as a preset threshold for the number of loops of 1000, then after the number of loops reaches 1000, stop training the semantic analysis model, even if the loss value exceeds the maximum threshold, use the model parameters obtained by the last adjustment as the parameters of the final semantic analysis model to obtain a trained semantic analysis model. Based on the method provided in this example, complete the training of the semantic analysis model to obtain a trained semantic analysis model.

[0140] Optional, Figure 7 The process of the customer complaint information processing method provided in the embodiment of the present application Figure 6 ,like Figure 7 As shown, it also includes:

[0141] S701: Receive feedback information sent by a user terminal, wherein the feedback information is generated by the user terminal in response to a user's feedback operation on a solution.

[0142] Combined with the scenario example, after obtaining a solution, the user can provide feedback on the solution, such as whether the solution is effective, accurate, and the efficiency of providing the solution, etc. The user enters feedback information about the solution on the user side, and the user side sends the feedback information to the server side.

[0143] S702: Count the values ​​of various indicators in the feedback information within a preset time period.

[0144] Combined with the scenario example, the preset time period can be determined according to the actual situation, which can be one month or one week. If the preset time period is one month, the feedback information received in that month can be counted at the end of each month, and the effectiveness and accuracy of the solution, as well as the efficiency of the solution, etc. can be counted from the feedback information. For example, for the indicator of whether the solution is effective, the effective ratio can be counted, for the indicator of whether the solution is accurate, the accurate ratio can be counted, and for the indicator of the efficiency of the solution, the high efficiency ratio can be counted.

[0145] S703: If the value of the indicator is lower than the preset value, the sentiment analysis model and the semantic analysis model are optimized and trained.

[0146] Combined with the scenario example, if the preset values ​​of the effective solution ratio, the accurate solution ratio, or the high efficiency solution ratio are all set to 80%, if at least one of the effective solution ratio, the accurate solution ratio, or the high efficiency solution ratio is lower than 80%, then it is determined that the sentiment analysis model and the semantic analysis model have prediction defects, and the sentiment analysis model and the semantic analysis model need to be optimized and trained in a targeted manner. Based on the method provided in this example, it is possible to determine whether the sentiment analysis model and the semantic analysis model are qualified based on specific indicators, and achieve the purpose of optimizing the sentiment analysis model and the semantic analysis model.

[0147] This embodiment first receives the customer complaint information sent by the user end, preprocesses the text information in the customer complaint information into text information in a preset format, extracts features from the user information to obtain corresponding user feature information, then inputs the text information in the preset format and the corresponding user feature information into the sentiment analysis model to obtain the corresponding user sentiment tendency, determines the processing priority of the customer complaint information by the user sentiment tendency, adds the customer complaint information to the queue to be processed, and at a preset specified time, inputs the corresponding text information of all the customer complaint information in the queue to be processed into the semantic analysis model in order according to the processing priority to obtain the corresponding solution, and finally sends the obtained solution to the user end so that the user end displays the solution. Therefore, based on the method provided in this embodiment, customer complaint information with a high processing priority can be processed first to improve the flexibility of processing customer complaint information.

[0148] Figure 8 This is a schematic diagram of the structure of a customer complaint information processing device provided in an embodiment of the present application. The customer complaint information processing device is applied to a server, including:

[0149] A receiving module 81 is used to receive customer complaint information sent by a user terminal, wherein the customer complaint information is generated by the user terminal in response to a customer complaint operation of the user, and wherein the customer complaint information includes user information and text information;

[0150] The processing module 82 is used to pre-process the text information in the customer complaint information to obtain text information in a preset format;

[0151] The processing module 82 is further used to extract features from the user information in the customer complaint information to obtain corresponding user feature information;

[0152] An input module 83, used to input the text information in a preset format in the customer complaint information and the corresponding user feature information into a sentiment analysis model to output the corresponding user sentiment tendency;

[0153] The processing module 82 is further used to determine the processing priority of the customer complaint information based on the user emotional tendency corresponding to the customer complaint information;

[0154] The processing module 82 is further used to add the customer complaint information to a queue to be processed;

[0155] The input module 83 is further used to input the text information in the customer complaint information in the waiting queue into the semantic analysis model in order according to the processing priority at a preset designated time, so as to output a corresponding solution;

[0156] The processing module 83 is further configured to send the obtained solution to the user terminal so that the user terminal displays the solution.

[0157] Optionally, the processing module 83 is specifically used to input the text information in a preset format in the customer complaint information and the corresponding user feature information into the sentiment analysis model, so that the sentiment analysis model performs feature extraction on the text information to output a first feature vector corresponding to the text information, and performs feature extraction on the user feature information to output a second feature vector corresponding to the user feature information;

[0158] The processing module 83 is further configured to determine a first weight corresponding to the text information and a second weight corresponding to the user feature information;

[0159] The processing module 83 is further configured to add the product of the first feature vector and the first weight to the product of the second feature vector and the first weight to obtain a fused feature vector;

[0160] The processing module 83 is further specifically used to input the fused feature vector into the sentiment analysis model to output the corresponding sentiment tendency.

[0161] Optionally, the processing module 83 is further configured to identify the type of the user's emotional tendency, wherein the type includes positive tendency, negative tendency and neutral tendency;

[0162] The processing module 83 is further configured to determine that the processing priority of the customer complaint information is the first processing priority if the user's emotional tendency is positive;

[0163] The processing module 83 is further configured to determine the processing priority of the customer complaint information as the second processing priority if the user's emotional tendency is a neutral tendency;

[0164] The processing module 83 is further specifically used to determine that the processing priority of the customer complaint information is a third processing priority if the user's emotional tendency is a negative tendency, wherein the third processing priority is higher than the second processing priority, and the second processing priority is higher than the first processing priority.

[0165] Optionally, the customer complaint information includes the customer complaint time;

[0166] Correspondingly, the processing module 83 is further specifically used to input the customer complaint information into the semantic analysis model in order from the earliest to the latest customer complaint time if there are two or more customer complaint information with the same processing priority, so as to output corresponding solutions.

[0167] Optionally, the processing module 83 is further configured to obtain first training data, wherein the first training data includes historical customer complaint information and standard sentiment tendency;

[0168] The processing module 83 is further configured to input the historical customer complaint information into a sentiment analysis model so that the sentiment analysis model outputs a predicted sentiment tendency;

[0169] The processing module 83 is further used to compare the predicted emotional tendency with the standard emotional tendency based on the loss function to obtain a corresponding loss value;

[0170] The processing module 83 is further configured to adjust the parameters of the sentiment analysis model if the loss value exceeds a preset threshold;

[0171] The processing module 83 is specifically used to loop through the steps of inputting the historical customer complaint information into the sentiment analysis model and adjusting the parameters of the sentiment analysis model until the loss value between the predicted sentiment tendency output by the sentiment analysis model and the standard sentiment tendency does not exceed the preset threshold, or stops when the number of loops reaches the preset threshold, so as to obtain a trained sentiment analysis model.

[0172] Optionally, the processing module 83 is further configured to obtain second training data, wherein the second training data includes historical customer complaint information and standard solutions;

[0173] The processing module 83 is further configured to input the historical customer complaint information into a semantic analysis model so that the semantic analysis model outputs a prediction solution;

[0174] The processing module 83 is further configured to compare the predicted solution with the standard solution based on the loss function to obtain a corresponding loss value;

[0175] The processing module 83 is further configured to adjust the parameters of the semantic analysis model based on a back propagation algorithm if the loss value exceeds a preset threshold;

[0176] The processing module 83 is specifically used to loop through the steps of inputting the historical customer complaint information into the semantic analysis model and adjusting the parameters of the semantic analysis model based on the back propagation algorithm, until the loss value between the predicted solution output by the semantic analysis model and the standard solution does not exceed the preset threshold, or stops when the number of cycles reaches the preset threshold, so as to obtain a trained semantic analysis model.

[0177] Optionally, the processing module 83 is further configured to receive feedback information sent by a user terminal, wherein the feedback information is generated by the user terminal in response to a user's feedback operation on a solution;

[0178] The processing module 83 is further used to count the values ​​of various indicators in the feedback information within a preset time period;

[0179] The processing module 83 is also specifically used to optimize the sentiment analysis model and the semantic analysis model if the value of the indicator is lower than a preset value.

[0180] The customer complaint information processing device provided in the embodiment of the present application can be used to execute the technical solution of the customer complaint information processing method in the above embodiment. Its implementation principle and technical effect are similar and will not be repeated here.

[0181] It should be noted that it should be understood that the division of the various modules of the above device is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And these modules can all be implemented in the form of software calling through processing elements; they can also be all implemented in the form of hardware; some modules can also be implemented in the form of software calling through processing elements, and some modules can be implemented in the form of hardware. For example, the processing module 83 can be a separately established processing element, or it can be integrated in a chip of the above device. In addition, it can also be stored in the memory of the above device in the form of program code, and called and executed by a processing element of the above device. The function of the above processing module 83. The implementation of other modules is similar. In addition, all or part of these modules can be integrated together, or they can be implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the processor element or instructions in the form of software.

[0182] Fig. 9 This is a schematic diagram of the structure of an electronic device provided in an embodiment of the present application. Fig. 9 As shown, the electronic device may include: a transceiver 121 , a processor 122 , and a memory 123 .

[0183] The processor 122 executes the computer execution instructions stored in the memory, so that the processor 122 executes the scheme in the above embodiment. The processor 122 can be a general-purpose processor, including a central processing unit CPU, a network processor (network processor, NP), etc.; it can also be a digital signal processor DSP, an application-specific integrated circuit ASIC, a field programmable gate array FPGA or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components.

[0184] The memory 123 is connected to the processor 122 via a system bus and completes communication between them. The memory 123 is used to store computer program instructions.

[0185] The transceiver 121 may be used to obtain tasks to be executed and configuration information of the tasks to be executed.

[0186] The system bus can be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus, etc. The system bus can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. The transceiver is used to realize the communication between the database access device and other computers (such as clients, read-write libraries, and read-only libraries). The memory may include random access memory (RAM) and may also include non-volatile memory.

[0187] The electronic device provided in the embodiment of the present application may be the terminal device of the above embodiment.

[0188] An embodiment of the present application also provides a chip for running instructions, which is used to execute the technical solution of the customer complaint information processing method in the above embodiment.

[0189] The embodiment of the present application further provides a computer-readable storage medium, in which computer instructions are stored. When the computer instructions are executed on a computer, the computer executes the technical solution of the customer complaint information processing method of the above embodiment.

[0190] An embodiment of the present application also provides a computer program product, which includes a computer program stored in a computer-readable storage medium. At least one processor can read the computer program from the computer-readable storage medium. When at least one processor executes the computer program, the technical solution of the customer complaint information processing method in the above embodiment can be implemented.

[0191] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic, for example, the division of modules is only a logical function division, and there may be other division methods in actual implementation, such as multiple modules can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or modules, which can be electrical, mechanical or other forms.

[0192] The modules described as separate components may or may not be physically separated, and the components shown as modules may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the modules may be selected according to actual needs to implement the solution of this embodiment.

[0193] In addition, each functional module in each embodiment of the present application can be integrated into one processing unit, or each module can exist physically separately, or two or more modules can be integrated into one unit. The above-mentioned module-composed unit can be implemented in the form of hardware or in the form of hardware plus software functional units.

[0194] The above-mentioned integrated module implemented in the form of a software function module can be stored in a computer-readable storage medium. The above-mentioned software function module is stored in a storage medium, including a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) or a processor to perform some steps of the methods of various embodiments of the present application.

[0195] It should be understood that the processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), etc. A general-purpose processor may be a microprocessor or any conventional processor. The steps of the method disclosed in the invention may be directly implemented as being executed by a hardware processor, or may be implemented by a combination of hardware and software modules in the processor.

[0196] The memory may include a high-speed RAM memory, and may also include a non-volatile storage NVM, such as at least one disk memory, and may also be a USB flash drive, a mobile hard disk, a read-only memory, a magnetic disk or an optical disk, etc.

[0197] The bus may be an Industry Standard Architecture (ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (EISA) bus, etc. The bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, the bus in the drawings of the present application is not limited to only one bus or one type of bus.

[0198] The above storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk. The storage medium can be any available medium that can be accessed by a general or special purpose computer.

[0199] An exemplary storage medium is coupled to a processor so that the processor can read information from the storage medium and write information to the storage medium. Of course, the storage medium can also be a component of the processor. The processor and the storage medium can be located in an application specific integrated circuit (ASIC). Of course, the processor and the storage medium can also exist as discrete components in an electronic control unit or a main control device.

[0200] Those skilled in the art can understand that all or part of the steps of implementing the above-mentioned method embodiments can be completed by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the aforementioned storage medium includes: ROM, RAM, disk or optical disk and other media that can store program codes.

[0201] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit it. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned 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 application.

Claims

1. A method for processing customer complaint information, characterized in that: Applied to the server, including: Receiving customer complaint information sent by a user terminal, wherein the customer complaint information is generated by the user terminal in response to a customer complaint operation of the user, and wherein the customer complaint information includes user information and text information; Preprocessing the text information in the customer complaint information to obtain text information in a preset format; Extracting features from the user information in the customer complaint information to obtain corresponding user feature information; Inputting the text information in a preset format in the customer complaint information and the corresponding user feature information into a sentiment analysis model to output the corresponding user sentiment tendency; Determining a processing priority of the customer complaint information based on the user sentiment tendency corresponding to the customer complaint information; Add the customer complaint information to a queue to be processed; At a preset designated time, the text information in the customer complaint information in the queue to be processed is input into the semantic analysis model in order according to the processing priority, so as to output the corresponding solution; The obtained solution is sent to the user end so that the user end displays the solution.

2. The method according to claim 1, characterized in that The step of inputting the text information in a preset format in the customer complaint information and the corresponding user feature information into a sentiment analysis model to output the corresponding user sentiment tendency includes: Inputting the text information in a preset format in the customer complaint information and the corresponding user feature information into a sentiment analysis model, so that the sentiment analysis model performs feature extraction on the text information to output a first feature vector corresponding to the text information, and performs feature extraction on the user feature information to output a second feature vector corresponding to the user feature information; Determine a first weight corresponding to the text information and a second weight corresponding to the user characteristic information; Adding the product of the first feature vector and the first weight to the product of the second feature vector and the first weight to obtain a fused feature vector; The fused feature vector is input into the sentiment analysis model to output the corresponding sentiment tendency.

3. The method according to claim 1, characterized in that The determining the processing priority of the customer complaint information based on the user emotional tendency corresponding to the customer complaint information includes: Identifying the type of the user's emotional tendency, wherein the type includes positive tendency, negative tendency and neutral tendency; If the user's emotional tendency is positive, determining the processing priority of the customer complaint information to be the first processing priority; If the user's emotional tendency is neutral, determining the processing priority of the customer complaint information to be the second processing priority; If the user's emotional tendency is negative, the processing priority of the customer complaint information is determined to be a third processing priority, wherein the third processing priority is higher than the second processing priority, and the second processing priority is higher than the first processing priority.

4. The method according to claim 1, characterized in that: The customer complaint information includes the customer complaint time; Accordingly, after determining the processing priority of the customer complaint information based on the user emotional tendency corresponding to the customer complaint information, the method further includes: If there are two or more customer complaints with the same processing priority, the customer complaint information will be input into the semantic analysis model in order from the earliest to the latest time of the customer complaint to output the corresponding solution.

5. The method according to claim 1, characterized in that The training process of the sentiment analysis model includes: Acquiring first training data, wherein the first training data includes historical customer complaint information and standard sentiment orientation; Inputting the historical customer complaint information into a sentiment analysis model so that the sentiment analysis model outputs a predicted sentiment tendency; Based on the loss function, the predicted sentiment tendency is compared with the standard sentiment tendency to obtain the corresponding loss value; If the loss value exceeds a preset threshold, adjusting the parameters of the sentiment analysis model; The steps from inputting the historical customer complaint information into the sentiment analysis model to adjusting the parameters of the sentiment analysis model are executed in a loop until the loss value between the predicted sentiment tendency output by the sentiment analysis model and the standard sentiment tendency does not exceed the preset threshold, or the loop is stopped when the number of times reaches the preset threshold, so as to obtain a trained sentiment analysis model.

6. The method according to claim 1, characterized in that The training process of the semantic analysis model includes: Acquiring second training data, wherein the second training data includes historical customer complaint information and standard solutions; Inputting the historical customer complaint information into a semantic analysis model so that the semantic analysis model outputs a prediction solution; Compare the predicted solution with the standard solution based on the loss function to obtain the corresponding loss value; If the loss value exceeds a preset threshold, adjusting the parameters of the semantic analysis model based on a back propagation algorithm; The steps from inputting the historical customer complaint information into the semantic analysis model to adjusting the parameters of the semantic analysis model based on the back propagation algorithm are executed cyclically until the loss value between the predicted solution output by the semantic analysis model and the standard solution does not exceed the preset threshold, or the cycle is stopped when the number of cycles reaches the preset threshold, so as to obtain a trained semantic analysis model.

7. The method according to any one of claims 1 to 6, characterized in that: Also includes: Receiving feedback information sent by a user terminal, wherein the feedback information is generated by the user terminal in response to a user's feedback operation on a solution; Count the values ​​of various indicators in the feedback information within a preset time period; If the value of the indicator is lower than the preset value, the sentiment analysis model and the semantic analysis model will be optimized and trained.

8. A customer complaint information processing device, characterized in that: Applied to the server, including: A receiving module, configured to receive customer complaint information sent by a user terminal, wherein the customer complaint information is generated by the user terminal in response to a customer complaint operation of the user, and wherein the customer complaint information includes user information and text information; A processing module, used for preprocessing the text information in the customer complaint information to obtain text information in a preset format; The processing module is further used to extract features from the user information in the customer complaint information to obtain corresponding user feature information; An input module, used to input the text information in a preset format in the customer complaint information and the corresponding user feature information into the sentiment analysis model to output the corresponding user sentiment tendency; The processing module is further used to determine the processing priority of the customer complaint information based on the user emotional tendency corresponding to the customer complaint information; The processing module is also used to add the customer complaint information to a queue to be processed; The input module is further used to input the text information in the customer complaint information in the queue to be processed into the semantic analysis model in sequence according to the processing priority at a preset specified time, so as to output a corresponding solution; The processing module is further used to send the obtained solution to the user terminal so that the user terminal displays the solution.

9. An electronic device, characterized in that: include: A processor, and a memory communicatively connected to the processor; The memory stores computer-executable instructions; The processor executes the computer-executable instructions stored in the memory to implement the method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer-executable instructions, which are used to implement the method according to any one of claims 1 to 7 when executed by a processor.

11. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 7 when being executed by a processor.

Images