Demand processing method and device based on emotion perception and electronic equipment

Abstract

The invention provides a demand processing method and device based on emotion perception and electronic equipment. The method comprises the following steps: acquiring user interaction information; basic information of the user is determined according to the user interaction information, the basic information of the user is analyzed through a singular value decomposition model, the user demand type is obtained, the singular value decomposition model is obtained through training of multiple sets of data, and each set of data in the multiple sets of data comprises historical basic information of the user and label data of the user demand type; the user basic information at least comprises a user name, and the user demand type comprises any one of consultation, complaint, suggestion and report; and determining a user emotion level according to the audio information in the user interaction information, and generating early warning information when the user emotion level is the first level so as to indicate a worker to perform demand processing according to the user demand type. According to the invention, the problem that an electric power system cannot accurately recognize the emotion of a user in the prior art is solved.

Description

Technical Field

[0001] The present application relates to the technical field of emotion perception, and in particular to a demand processing method, device, computer-readable storage medium and electronic device based on emotion perception. Background Art

[0002] Customer service centers play a key role in helping companies improve customer service experience and enhance brand influence. With the advancement of digital transformation, customers' demand for quick response and personalized services is growing. Therefore, an efficient emotion-aware customer service system has become the key to improving customer satisfaction and corporate competitiveness. Traditional customer service systems mainly rely on manual processing, which is inefficient and difficult to provide a consistent customer service experience. Therefore, it has become a trend to use artificial intelligence technology to build an automated emotion-aware customer service system to achieve emotion recognition, automatic warning and intervention to improve customer service efficiency and satisfaction.

[0003] Existing emotion-aware customer service systems mainly use deep learning algorithms to perform emotion analysis on multimodal data such as user voice and text. However, these systems still face some challenges in practical applications:

[0004] 1. Low data quality: Since customer service call recordings may contain noise, unclear voices, and other issues, it is difficult to obtain high-quality data during model training, which affects the accuracy of emotion recognition.

[0005] 2. Limited accuracy in emotion recognition: For the recognition of complex emotions, such as subtle emotional changes, background noise, etc., existing deep learning models may find it difficult to accurately capture and analyze them, resulting in limited recognition accuracy.

[0006] 3. Low efficiency of customer service ticket classification: Based on the results of sentiment analysis, automatic ticket classification is also a key link to improve customer service efficiency. However, the current classification method may not fully consider the specific demands and emotional fluctuations of users, resulting in low accuracy of ticket classification.

[0007] 4. Insufficient security in information transmission: In actual applications, the customer service system needs to process a large amount of sensitive user information, such as call recordings, personal information, etc. If the security of information transmission is not guaranteed, it may cause security risks such as data leakage.

[0008] Therefore, how to improve the data quality, emotion recognition accuracy, automatic work order classification efficiency and information transmission security of the customer service emotion perception system has become an urgent problem to be solved. Summary of the invention

[0009] The main purpose of the present application is to provide a demand processing method, device, computer-readable storage medium and electronic device based on emotion perception, so as to at least solve the problem in the prior art that the power system cannot accurately identify the user's emotions.

[0010] In order to achieve the above-mentioned purpose, according to one aspect of the present application, a demand processing method based on emotion perception is provided, including: obtaining user interaction information, wherein the user interaction information at least includes audio information sent by the user and representing the user demand; determining the user basic information according to the user interaction information, and analyzing the user basic information through a singular value decomposition model to obtain the user demand type, wherein the singular value decomposition model is obtained by training multiple groups of data, each group of the multiple groups of data includes historical user basic information and label data of user demand type, the user basic information at least includes the user name, and the user demand type includes any one of consultation, complaint, suggestion and report; determining the user emotion level according to the audio information in the user interaction information, and when the user emotion level is level one, generating early warning information to instruct the staff to process the demand according to the user demand type, wherein the user emotion level of level one indicates that the user is in a bad mood.

[0011] Optionally, determining the user emotion level according to the audio information in the user interaction information includes: obtaining the decibel of the audio information, and constructing a one-to-one mapping relationship between the decibel and the corresponding moment; determining the decibel average value according to the one-to-one mapping relationship, and determining that the user emotion is level one when the decibel average value is greater than or equal to a preset threshold.

[0012] Optionally, determining the user's basic information based on the user interaction information includes: receiving identity authentication information sent by the user, and verifying whether the identity authentication information is consistent with pre-stored information, wherein the identity authentication information is sent by the user in response to an identity authentication request, and the pre-stored information is identity information pre-stored by the user; when the identity authentication information is consistent with the pre-stored information, obtaining the basic information of the user.

[0013] Optionally, before analyzing the user basic information through a singular value decomposition model to obtain the user demand type, the method also includes: obtaining historical user basic information and performing data cleaning on the historical user basic information; marking features representing the user demand type in the historical user basic information after data cleaning to obtain the label data representing the user demand type.

[0014] Optionally, the user basic information is analyzed through a singular value decomposition model to obtain the user demand type, including: constructing a sparse matrix of the user basic information and determining the initial value of the sparse matrix; constructing an objective function according to the sparse matrix, solving the objective function through parameter iteration, and obtaining the user demand type.

[0015] Optionally, after generating the warning information, the method further includes: sending the warning information to the staff, so that the staff makes a return visit and calms the user's emotions.

[0016] Optionally, the method further includes: storing the user interaction information, the user basic information and the user demand type in a preset knowledge base for use in optimizing the singular value decomposition model.

[0017] According to another aspect of the present application, a demand processing device based on emotion perception is provided, including: an acquisition unit, used to acquire user interaction information, wherein the user interaction information at least includes audio information sent by the user and representing the user demand; an analysis unit, used to determine user basic information based on the user interaction information, and analyze the user basic information through a singular value decomposition model to obtain the user demand type, wherein the singular value decomposition model is obtained by training multiple groups of data, each group of the multiple groups of data includes historical user basic information and label data of user demand type, the user basic information at least includes the user name, and the user demand type includes any one of consultation, complaint, suggestion and report; a generation unit, used to determine the user emotion level based on the audio information in the user interaction information, and when the user emotion level is level one, generate warning information to instruct the staff to process the demand according to the user demand type, wherein the user emotion level of level one indicates that the user emotion is poor.

[0018] According to another aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute any one of the demand processing methods based on emotion perception.

[0019] According to another aspect of the present application, an electronic device is provided, comprising: one or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing any one of the emotion-awareness-based demand processing methods.

[0020] The technical solution of the present application is applied to obtain user interaction information, which at least includes audio information sent by the user to represent the user's needs; determine the user's basic information according to the user interaction information, and analyze the user's basic information through a singular value decomposition model to obtain the user's demand type, the singular value decomposition model is obtained by training multiple groups of data, each of the multiple groups of data includes historical user basic information and label data of the user's demand type, the user's basic information at least includes the user's name, and the user's demand type includes any one of consultation, complaint, suggestion and report; determine the user's emotional level according to the audio information in the user interaction information, and generate warning information when the user's emotional level is level one to instruct the staff to process the demand according to the user's demand type, and the user's emotional level is level one, indicating that the user's emotion is poor. Compared with the prior art, which cannot effectively identify user emotions, the present application first determines the type of user demand through a singular value decomposition model, and then determines the user's emotion through audio information, and finally generates warning information, so that the customer's demand can be processed according to the customer's emotion to meet the user's demand. Therefore, it can solve the problem that the power system in the prior art cannot accurately identify the user's emotions, and achieve the effect of accurately perceiving the user's emotions. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The drawings constituting part of the present application are used to provide a further understanding of the present application. The exemplary embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0022] Figure 1 A hardware structure block diagram of a mobile terminal for executing a demand processing method based on emotion perception provided by an embodiment of the present application is shown;

[0023] Figure 2 A schematic diagram of a flow chart of a demand processing method based on emotion perception provided by an embodiment of the present application is shown;

[0024] Figure 3 A structural block diagram of a demand processing device based on emotion perception provided in an embodiment of the present application is shown.

[0025] The above drawings include the following reference numerals:

[0026] 102, processor; 104, memory; 106, transmission device; 108, input and output devices. DETAILED DESCRIPTION

[0027] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0028] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0029] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequential order. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present application described here. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0030] For the convenience of description, some nouns or terms involved in the embodiments of the present application are explained below:

[0031] LC-SVD (Locality-Sensitive Hashing for Approximate Nearest Neighbor Search) is a locality-sensitive hashing algorithm for approximate nearest neighbor search. It is implemented by calculating the locality-sensitive hash values ​​of data points.

[0032] As introduced in the background technology, the emotion-aware customer service system in the prior art has problems such as low data quality, limited emotion recognition accuracy, and low customer service ticket classification efficiency, and thus cannot accurately identify user emotions. In order to solve the problem of not being able to accurately identify user emotions, the embodiments of the present application provide a demand processing method, device, computer-readable storage medium, and electronic device based on emotion perception.

[0033] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0034] The method embodiments provided in the embodiments of the present application can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 is a hardware structure block diagram of a mobile terminal of a demand processing method based on emotion perception according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0035] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the demand processing method based on emotion perception in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, the above method is implemented. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal 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 a combination thereof. The transmission device 106 is used to receive or send data via a network. The above-mentioned specific examples of the network may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0036] In this embodiment, a demand processing method based on emotion perception is provided, which runs on a mobile terminal, a computer terminal or a similar computing device. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although the logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0037] Figure 2is a flow chart of a method for processing demand based on emotion perception according to an embodiment of the present application. Figure 2 As shown, the method comprises the following steps:

[0038] Step S201, obtaining user interaction information, wherein the user interaction information at least includes audio information sent by the user and representing the user's needs;

[0039] Specifically, the system first needs to obtain user interaction information, which includes but is not limited to text, voice or audio information sent by users through phone calls, instant messages, emails or online forms, etc. It is particularly emphasized that audio information is a key data source for the system to recognize user emotions, because voice characteristics such as pitch, speaking speed, pauses, etc. can provide rich non-verbal emotional clues.

[0040] Step S202, determining basic user information according to the user interaction information, and analyzing the basic user information through a singular value decomposition model to obtain a user demand type, wherein the singular value decomposition model is obtained by training multiple sets of data, each set of the multiple sets of data includes historical user basic information and label data of user demand type, the user basic information includes at least a user name, and the user demand type includes any one of consultation, complaint, suggestion and report;

[0041] Specifically, after collecting the basic information of users, a user information database is constructed, and then the singular value decomposition model is trained. The system uses the historical basic information of users and their demand type labels to train through the singular value decomposition (SVD) model. SVD is a powerful matrix decomposition technology that can extract low-dimensional feature vectors from high-dimensional data, thereby simplifying the data processing process. During the training process, the model learns the intrinsic connection between the basic information of users and the demand type, establishes a user demand type prediction model, and performs real-time prediction of user demand types.

[0042] Step S203, determining the user emotion level according to the audio information in the user interaction information, and generating warning information when the user emotion level is level one to instruct the staff to handle the demand according to the user demand type, wherein the user emotion level being level one indicates that the user is in a bad mood.

[0043] Specifically, once the system determines that the user's emotional level has reached level one, an early warning message is immediately generated. The early warning message not only includes an assessment of the user's emotional state, but also includes the identification results of the user's demand type (inquiry, complaint, suggestion, report). The system sends the early warning message to the corresponding customer service or management personnel's device through SMS, email, or application notification to ensure a quick response. Staff who receive the early warning message should give priority to user requests with an emotional level of level one. During the processing process, the staff takes measures such as soothing, explaining, and responding quickly to promptly solve user problems, alleviate their negative emotions, and avoid escalation of complaints.

[0044] Through this embodiment, user interaction information is obtained, and the user interaction information at least includes audio information sent by the user to characterize the user's needs; the user's basic information is determined according to the user interaction information, and the user's basic information is analyzed by a singular value decomposition model to obtain the user's demand type, and the singular value decomposition model is obtained by training multiple groups of data, each of which includes historical user basic information and label data of user demand type, and the user's basic information at least includes the user's name, and the user demand type includes any one of consultation, complaint, suggestion and report; the user's emotional level is determined according to the audio information in the user interaction information, and when the user's emotional level is level one, an early warning message is generated to instruct the staff to process the demand according to the user's demand type, and the user's emotional level is level one, indicating that the user's emotion is poor. Compared with the prior art, which cannot effectively identify user emotions, the present application first determines the type of user demand through a singular value decomposition model, and then determines the user's emotion through audio information, and finally generates an early warning message, so that the customer's demand can be processed according to the customer's emotion to meet the user's demand. Therefore, the problem that the power system in the prior art cannot accurately identify the user's emotions can be solved, and the effect of accurately perceiving the user's emotions can be achieved.

[0045] In the specific implementation process, the above step S203 determines the user's emotional level according to the audio information in the user interaction information, which can be achieved through the following steps: Step S2031: Obtain the decibel of the audio information, and construct a one-to-one mapping relationship between the decibel and the corresponding time; Step S2032: Determine the decibel average value according to the one-to-one mapping relationship, and determine that the user's emotion is level 1 when the decibel average value is greater than or equal to a preset threshold. The method determines the user level through the above steps, so that emotional evaluation and early warning triggering can be performed based on volume changes, which can effectively improve customer service efficiency and user satisfaction.

[0046] Specifically, firstly, audio information is captured from user interaction, and the audio processing module is used to measure the decibel level of the audio in real time. The decibel value can reflect the volume of the user's voice and is one of the important indicators for evaluating the user's emotional state. The audio decibel value at each moment is recorded, and a mapping relationship between decibel and time is established. This mapping relationship can reflect the changing trend of user emotions over time and provide time series data for subsequent emotional evaluation. The decibel values ​​within a certain call or interaction time period are averaged to obtain the decibel average value. The preset threshold is obtained based on a large amount of historical data statistical analysis and is usually associated with the user's emotional change pattern. When the decibel average reaches or exceeds the preset threshold, the system determines that the user's emotion may be in a highly excited or dissatisfied state. If the calculated decibel average is greater than or equal to the preset threshold, the system automatically evaluates the user's emotional level as level one, indicating that the user is in a bad mood. At this point, the system immediately starts the early warning mechanism and generates an early warning message, prompting customer service personnel or management personnel to intervene in time and pay attention to the user's request to avoid complaints or negative impacts caused by emotional escalation. In addition to the decibel average, other audio features such as speech speed, pauses, and pitch changes, as well as sensitive word detection in text information, can also be combined to perform multi-dimensional emotional analysis. This comprehensive analysis method can provide a more comprehensive sentiment assessment and further improve the sensitivity and reliability of the early warning mechanism.

[0047] In some optional implementations, the above step S202 determines the basic information of the user according to the user interaction information, which can be implemented by the following steps: Step S2021: receiving the identity authentication information sent by the user, and verifying whether the identity authentication information is consistent with the pre-stored information, wherein the identity authentication information is sent by the user in response to the identity authentication request, and the pre-stored information is the identity information pre-stored by the user; Step S2022: when the identity authentication information is consistent with the pre-stored information, obtaining the basic information of the user. This method can not only ensure the security and personalization of the service, but also effectively improve the customer service efficiency and enhance the user's trust and satisfaction with the service.

[0048] In the specific implementation process, when a user attempts to access or use certain functions of the customer service system, the system first initiates an identity authentication request and requires the user to provide information that can prove his or her identity. This step is intended to ensure the security and fairness of the service and prevent unauthorized access. After receiving the identity authentication request, the user provides identity authentication information. This information may include but is not limited to the user's name, ID number, mobile phone number, registered email address, facial recognition image, voiceprint recognition sample, etc., depending on the security level and verification method designed by the system. The system has a set of pre-stored information databases, which contain the identity information pre-submitted and verified by the user. The system compares the received identity authentication information with the information in the pre-stored database to verify the authenticity of the user's identity. In order to improve the accuracy and efficiency of identity authentication, the system adopts an efficient data comparison algorithm. The algorithm can quickly identify and compare the identity authentication information and pre-stored information provided by the user, and can quickly obtain verification results even in a massive data environment. If the system determines that the identity authentication information provided by the user is consistent with the pre-stored information, that is, it meets the preset verification standard, the system will consider that the user's identity authentication is successful. At this time, the system further obtains the user's detailed basic information to provide data support for subsequent services. Once the identity authentication is successful, the system will control its access to specific services or functions based on the user's identity and permissions. For example, for VIP users, the customer service system can provide priority service, and for ordinary users, the system handles requests according to standard procedures.

[0049] In some other optional implementations, before analyzing the basic user information through the singular value decomposition model to obtain the user demand type, the method further includes: step S204: obtaining historical user basic information and performing data cleaning on the historical user basic information; step S205: labeling the features representing the user demand type in the historical user basic information after data cleaning to obtain the label data representing the user demand type. This method not only provides a high-quality data foundation for demand type prediction, but also ensures the reliability and cleanliness of the data, creating conditions for the efficient operation of the system.

[0050] In the specific implementation process, data cleaning is a key step in data preprocessing, which aims to remove invalid information, fill missing values, eliminate duplicate records, filter abnormal data, etc. Specifically, it includes: Invalid information removal: exclude noise information irrelevant to user needs, such as background noise, irrelevant conversations, etc. Missing value filling: use statistical methods or machine learning prediction to fill in the missing parts in user information to ensure the integrity and availability of data. Abnormal data filtering: identify and exclude abnormal data points that obviously deviate from the normal range to avoid bias in the analysis results. Duplicate record elimination: identify and merge duplicate user records through data comparison to ensure the uniqueness and accuracy of each user's data. The system extracts features from the historical user interaction information after data cleaning and identifies features related to the type of user needs. For example, extract keywords, conversation duration, call frequency, etc. from user call records, and extract emotional tendencies and the frequency of occurrence of sensitive words from user text messages. Use natural language processing technology to identify keywords in user interaction texts, such as "complaints", "consultation", "suggestions", etc. These keywords are directly related to the type of user needs. Through sentiment analysis algorithms, we can determine the emotional tendencies contained in the information of user interactions, such as positive, negative, and neutral, which helps to further understand the background and motivation of user needs.

[0051] In some optional implementations, the above step S202 analyzes the basic information of the user through a singular value decomposition model to obtain the user demand type, which can be achieved by the following steps: step S2023: construct a sparse matrix of the basic information of the user and determine the initial value of the sparse matrix; step S2024: construct an objective function according to the sparse matrix, solve the objective function through parameter iteration, and obtain the user demand type. This method combines the prediction method of sparse matrix representation and iterative optimization algorithm, which can not only effectively process high-dimensional information, but also improve the accuracy and efficiency of demand type prediction.

[0052] Specifically, sparse matrices are an efficient way to store and process high-dimensional sparse data. Perform matrix initialization and determine the initial value. The objective function is a key component used to guide model training and parameter optimization in machine learning and data mining. In this scenario, the construction of the objective function needs to consider the following factors: The objective function should be able to reflect the relationship between the user's demand type and the user's basic information, and maximize the consistency between the predicted demand type and the actual demand type through matrix operations and parameter adjustments. In order to fully utilize the advantages of sparse matrices, the objective function must also include terms for sparse reconstruction errors so that the model can maintain high prediction accuracy on sparse matrices. The discriminant term is introduced to enhance the classification ability of the model, and the accuracy and stability of demand type prediction are improved by maximizing the differences between different demand types. The constructed objective function is solved by an iterative optimization method to obtain the optimal parameter configuration.

[0053] In some optional real-time methods, after generating the warning information, the method further includes the following steps: Step S206: sending the warning information to the staff so that the staff can make a return visit and appease the user. This method can improve user experience, prevent complaints from escalating, and improve customer service efficiency.

[0054] Specifically, the warning information should be designed in a concise and clear format while containing sufficient information. The information can be sent to the customer service staff in the form of SMS, email, application notification or phone call. The content design should include: Basic user information: such as user ID, name, contact information, etc. Emotional state description: clear emotional level identification, and a brief description of the emotional state. User demand overview: provide a specific content summary of user consultation, complaints, suggestions or reports. Handling suggestions: provide preliminary handling suggestions or action guidelines based on preset processes or model predictions. Intelligently match the most suitable customer service staff to receive warning information based on factors such as work schedules, professional skills, and user demand types. During non-working hours, the information can be sent to the on-duty staff to ensure timely response. The staff who receive the warning information should contact the user immediately or in the shortest time for a return visit. During the return visit, the staff should adopt a series of emotional soothing strategies, including but not limited to: Listening and empathy: first listen patiently to the user's problems and dissatisfaction, express understanding and sympathy, and make the user feel valued. Explanation and clarification: provide detailed explanations and clarifications for the user's problems to eliminate misunderstandings. Quick response: If the user's problem can be solved immediately, take prompt action to give the user a satisfactory answer. Escalation: For complex or difficult to solve problems immediately, they should be promptly escalated to a higher level of customer service or relevant departments to ensure that the problem is properly handled. Emotional soothing skills: Use professional emotional soothing skills, such as tone adjustment, emotional transfer, positive feedback, etc., to relieve user tension and improve service experience.

[0055] In some optional real-time modes, the method further includes step S207: storing the user interaction information, the user basic information and the user demand type in a preset knowledge base for optimizing the singular value decomposition model. This method can improve the accuracy and efficiency of model prediction and provide more accurate decision support for the system.

[0056] In the specific implementation process, the knowledge base should be designed as a database structure that can efficiently store and retrieve a large amount of user data. It includes but is not limited to data items such as user interaction history, user basic information, and demand type labels. The steps of knowledge base construction include: Database design: determine the data table structure, field type, relationship model, etc. to meet the storage requirements of different types of user data. Data import: import the pre-processed user data into the knowledge base to ensure the integrity and accuracy of the data. Index and optimization: create appropriate indexes, optimize query performance, and ensure the system's response speed when processing large amounts of data.

[0057] In order to enable those skilled in the art to more clearly understand the technical solution of the present application, the implementation process of the demand processing method based on emotion perception of the present application will be described in detail below in combination with specific embodiments.

[0058] This embodiment relates to a specific demand processing method based on emotion perception, including the following steps:

[0059] Step S1: Obtaining user's incoming call or text information;

[0060] Step S2: using the trained LC-SVD algorithm (singular value decomposition model) to process the above data, automatically predicting the demands or possible business classifications, i.e., labels or user historical demand information;

[0061] Step S3: predicting the user's emotion by the decibel of the user's voice;

[0062] Step S4: When the user's emotion level is level one, a warning message is generated and sent to the staff, and the work order business is automatically classified to prevent unreasonable demands;

[0063] Step S5: The staff processes the above work order in a timely manner.

[0064] The embodiment of the present application also provides a demand processing device based on emotion perception. It should be noted that the demand processing device based on emotion perception in the embodiment of the present application can be used to execute the demand processing method based on emotion perception provided in the embodiment of the present application. The device is used to implement the above-mentioned embodiments and preferred implementation modes, and the descriptions that have been made will not be repeated. As used below, the term "module" can implement a combination of software and / or hardware for a predetermined function. Although the device described in the following embodiments is preferably implemented in software, the implementation of hardware, or a combination of software and hardware, is also possible and conceived.

[0065] The following is an introduction to the demand processing device based on emotion perception provided in an embodiment of the present application.

[0066] Figure 3is a schematic diagram of a demand processing device based on emotion perception according to an embodiment of the present application. Figure 3 As shown, the device comprises:

[0067] The acquisition unit 10 is used to acquire user interaction information, wherein the user interaction information at least includes audio information sent by the user and representing the user's needs;

[0068] Specifically, the system first needs to obtain user interaction information, which includes but is not limited to text, voice or audio information sent by users through phone calls, instant messages, emails or online forms, etc. It is particularly emphasized that audio information is a key data source for the system to recognize user emotions, because voice characteristics such as pitch, speaking speed, pauses, etc. can provide rich non-verbal emotional clues.

[0069] An analysis unit 20 is used to determine basic user information according to the user interaction information, and analyze the basic user information through a singular value decomposition model to obtain a user demand type, wherein the singular value decomposition model is obtained by training multiple sets of data, each set of the multiple sets of data includes historical user basic information and label data of user demand type, the user basic information includes at least a user name, and the user demand type includes any one of consultation, complaint, suggestion and report;

[0070] Specifically, after collecting the basic information of users, a user information database is constructed, and then the singular value decomposition model is trained. The system uses the historical basic information of users and their demand type labels to train through the singular value decomposition (SVD) model. SVD is a powerful matrix decomposition technology that can extract low-dimensional feature vectors from high-dimensional data, thereby simplifying the data processing process. During the training process, the model learns the intrinsic connection between the basic information of users and the demand type, establishes a user demand type prediction model, and performs real-time prediction of user demand types.

[0071] The generating unit 30 is used to determine the user emotion level according to the audio information in the user interaction information, and generate warning information when the user emotion level is level one to instruct the staff to handle the demand according to the user demand type, wherein the user emotion level of level one indicates that the user is in a bad mood.

[0072] Specifically, once the system determines that the user's emotional level has reached level one, an early warning message is immediately generated. The early warning message not only includes an assessment of the user's emotional state, but also includes the identification results of the user's demand type (inquiry, complaint, suggestion, report). The system sends the early warning message to the corresponding customer service or management personnel's device through SMS, email, or application notification to ensure a quick response. Staff who receive the early warning message should give priority to user requests with an emotional level of level one. During the processing process, the staff takes measures such as soothing, explaining, and responding quickly to promptly solve user problems, alleviate their negative emotions, and avoid escalation of complaints.

[0073] Through this embodiment, user interaction information is obtained, and the user interaction information at least includes audio information sent by the user to characterize the user's needs; the user's basic information is determined according to the user interaction information, and the user's basic information is analyzed by a singular value decomposition model to obtain the user's demand type, and the singular value decomposition model is obtained by training multiple groups of data, each of which includes historical user basic information and label data of user demand type, and the user's basic information at least includes the user's name, and the user demand type includes any one of consultation, complaint, suggestion and report; the user's emotional level is determined according to the audio information in the user interaction information, and when the user's emotional level is level one, an early warning message is generated to instruct the staff to process the demand according to the user's demand type, and the user's emotional level is level one, indicating that the user's emotion is poor. Compared with the prior art, which cannot effectively identify user emotions, the present application first determines the type of user demand through a singular value decomposition model, and then determines the user's emotion through audio information, and finally generates an early warning message, so that the customer's demand can be processed according to the customer's emotion to meet the user's demand. Therefore, the problem that the power system in the prior art cannot accurately identify the user's emotions can be solved, and the effect of accurately perceiving the user's emotions can be achieved.

[0074] In the specific implementation process, the above-mentioned generation unit includes a construction module and a first determination module. The construction module is used to obtain the decibel of the audio information and construct a one-to-one mapping relationship between the decibel and the corresponding time; the first determination module is used to determine the decibel average value according to the one-to-one mapping relationship, and when the decibel average value is greater than or equal to a preset threshold, the user emotion is determined to be level one. The device determines the user level through the above steps, so that emotion evaluation and early warning triggering can be performed based on volume changes, which can effectively improve customer service efficiency and user satisfaction.

[0075] Specifically, firstly, audio information is captured from user interaction, and the audio processing module is used to measure the decibel level of the audio in real time. The decibel value can reflect the volume of the user's voice and is one of the important indicators for evaluating the user's emotional state. The audio decibel value at each moment is recorded, and a mapping relationship between decibel and time is established. This mapping relationship can reflect the changing trend of user emotions over time and provide time series data for subsequent emotional evaluation. The decibel values ​​within a certain call or interaction time period are averaged to obtain the decibel average value. The preset threshold is obtained based on a large amount of historical data statistical analysis and is usually associated with the user's emotional change pattern. When the decibel average reaches or exceeds the preset threshold, the system determines that the user's emotion may be in a highly excited or dissatisfied state. If the calculated decibel average is greater than or equal to the preset threshold, the system automatically evaluates the user's emotional level as level one, indicating that the user is in a bad mood. At this point, the system immediately starts the early warning mechanism and generates an early warning message, prompting customer service personnel or management personnel to intervene in time and pay attention to the user's request to avoid complaints or negative impacts caused by emotional escalation. In addition to the decibel average, other audio features such as speech speed, pauses, and pitch changes, as well as sensitive word detection in text information, can also be combined to perform multi-dimensional emotional analysis. This comprehensive analysis device can provide a more comprehensive emotional assessment, further improving the sensitivity and reliability of the early warning mechanism.

[0076] In some optional implementations, the analysis unit includes a verification module and an acquisition module, the verification module is used to receive identity authentication information sent by the user, and verify whether the identity authentication information is consistent with the pre-stored information, wherein the identity authentication information is sent by the user in response to the identity authentication request, and the pre-stored information is the identity information pre-stored by the user; the acquisition module is used to obtain the basic information of the user when the identity authentication information is consistent with the pre-stored information. The device can not only ensure the security and personalization of the service, but also effectively improve the customer service efficiency and enhance the user's trust and satisfaction with the service.

[0077] In the specific implementation process, when a user attempts to access or use certain functions of the customer service system, the system first initiates an identity authentication request and requires the user to provide information that can prove his or her identity. This step is intended to ensure the security and fairness of the service and prevent unauthorized access. After receiving the identity authentication request, the user provides identity authentication information. This information may include but is not limited to the user's name, ID number, mobile phone number, registered email address, facial recognition image, voiceprint recognition sample, etc., depending on the security level and verification method designed by the system. The system has a set of pre-stored information databases, which contain the identity information pre-submitted and verified by the user. The system compares the received identity authentication information with the information in the pre-stored database to verify the authenticity of the user's identity. In order to improve the accuracy and efficiency of identity authentication, the system adopts an efficient data comparison algorithm. The algorithm can quickly identify and compare the identity authentication information and pre-stored information provided by the user, and can quickly obtain verification results even in a massive data environment. If the system determines that the identity authentication information provided by the user is consistent with the pre-stored information, that is, it meets the preset verification standard, the system will consider that the user's identity authentication is successful. At this time, the system further obtains the user's detailed basic information to provide data support for subsequent services. Once the identity authentication is successful, the system will control its access to specific services or functions based on the user's identity and permissions. For example, for VIP users, the customer service system can provide priority service, and for ordinary users, the system handles requests according to standard procedures.

[0078] In some other optional implementations, the device further includes a data cleaning unit and a labeling unit. The data cleaning unit is used to obtain historical user basic information and perform data cleaning on the historical user basic information before analyzing the user basic information through a singular value decomposition model to obtain the user demand type; the labeling unit is used to label the features representing the user demand type in the historical user basic information after data cleaning to obtain the label data representing the user demand type. The device not only provides a high-quality data basis for demand type prediction, but also ensures the reliability and cleanliness of the data, creating conditions for the efficient operation of the system.

[0079] In the specific implementation process, data cleaning is a key step in data preprocessing, which aims to remove invalid information, fill missing values, eliminate duplicate records, filter abnormal data, etc. Specifically, it includes: Invalid information removal: exclude noise information irrelevant to user needs, such as background noise, irrelevant conversations, etc. Missing value filling: use statistical devices or machine learning prediction to fill in the missing parts in user information to ensure the integrity and availability of data. Abnormal data filtering: identify and exclude abnormal data points that obviously deviate from the normal range to avoid bias in the analysis results. Duplicate record elimination: through data comparison, identify and merge duplicate user records to ensure the uniqueness and accuracy of each user's data. The system extracts features from the historical user interaction information after data cleaning and identifies features related to the type of user needs. For example, extract keywords, conversation duration, call frequency, etc. from user call records, and extract emotional tendencies and the frequency of occurrence of sensitive words from user text messages. Use natural language processing technology to identify keywords in user interaction texts, such as "complaints", "consultation", "suggestions", etc. These keywords are directly related to the type of user needs. Through sentiment analysis algorithms, we can determine the emotional tendencies contained in the information of user interactions, such as positive, negative, and neutral, which helps to further understand the background and motivation of user needs.

[0080] In some optional implementations, the analysis unit includes a second determination module and a solution module, the second determination module is used to construct a sparse matrix of the user's basic information and determine the initial value of the sparse matrix; the solution module is used to construct an objective function according to the sparse matrix, solve the objective function through parameter iteration, and obtain the user demand type. The device combines a prediction device with a sparse matrix representation and an iterative optimization algorithm, which can not only effectively process high-dimensional information, but also improve the accuracy and efficiency of demand type prediction.

[0081] Specifically, a sparse matrix is ​​an efficient device for storing and processing high-dimensional sparse data. Perform matrix initialization and determine the initial value. The objective function is a key component used to guide model training and parameter optimization in machine learning and data mining. In this scenario, the construction of the objective function needs to consider the following factors: the objective function should be able to reflect the relationship between the user's demand type and the user's basic information, and maximize the consistency between the predicted demand type and the actual demand type through matrix operations and parameter adjustments. In order to fully utilize the advantages of sparse matrices, the objective function must also include terms for sparse reconstruction errors so that the model can maintain a high prediction accuracy on sparse matrices. The discriminant term is introduced to enhance the classification ability of the model, and the accuracy and stability of demand type prediction are improved by maximizing the differences between different demand types. The constructed objective function is solved by an iterative optimization device to obtain the optimal parameter configuration.

[0082] In some optional real-time methods, after generating the warning information, the device further includes a sending unit for sending the warning information to the staff, so that the staff can make a return visit and appease the user. The device can improve user experience, prevent complaints from escalating, and improve customer service efficiency.

[0083] Specifically, the warning information should be designed in a concise and clear format while containing sufficient information. The information can be sent to the customer service staff in the form of SMS, email, application notification or phone call. The content design should include: Basic user information: such as user ID, name, contact information, etc. Emotional state description: clear emotional level identification, and a brief description of the emotional state. User demand overview: provide a specific content summary of user consultation, complaints, suggestions or reports. Handling suggestions: provide preliminary handling suggestions or action guidelines based on preset processes or model predictions. Intelligently match the most suitable customer service staff to receive warning information based on factors such as work schedules, professional skills, and user demand types. During non-working hours, the information can be sent to the on-duty staff to ensure timely response. The staff who receive the warning information should contact the user immediately or in the shortest time for a return visit. During the return visit, the staff should adopt a series of emotional soothing strategies, including but not limited to: Listening and empathy: first listen patiently to the user's problems and dissatisfaction, express understanding and sympathy, and make the user feel valued. Explanation and clarification: provide detailed explanations and clarifications for the user's problems to eliminate misunderstandings. Quick response: If the user's problem can be solved immediately, take prompt action to give the user a satisfactory answer. Escalation: For complex or difficult to solve problems immediately, they should be promptly escalated to a higher level of customer service or relevant departments to ensure that the problem is properly handled. Emotional soothing skills: Use professional emotional soothing skills, such as tone adjustment, emotional transfer, positive feedback, etc., to relieve user tension and improve service experience.

[0084] In some optional real-time methods, the device further includes a storage unit for storing the user interaction information, the user basic information and the user demand type in a preset knowledge base for optimizing the singular value decomposition model. The device can improve the accuracy and efficiency of model prediction and provide more accurate decision support for the system.

[0085] In the specific implementation process, the knowledge base should be designed as a database structure that can efficiently store and retrieve a large amount of user data. It includes but is not limited to data items such as user interaction history, user basic information, and demand type labels. The steps of knowledge base construction include: Database design: determine the data table structure, field type, relationship model, etc. to meet the storage requirements of different types of user data. Data import: import the pre-processed user data into the knowledge base to ensure the integrity and accuracy of the data. Index and optimization: create appropriate indexes, optimize query performance, and ensure the system's response speed when processing large amounts of data.

[0086] The demand processing device based on emotion perception includes a processor and a memory. The acquisition unit, analysis unit, generation unit, etc. are all stored in the memory as program units, and the processor executes the program units stored in the memory to implement corresponding functions. The modules are all located in the same processor; or, the modules are located in different processors in any combination.

[0087] The processor includes a kernel, which calls the corresponding program unit from the memory. One or more kernels can be set, and the user's emotions can be accurately identified by adjusting the kernel parameters.

[0088] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM, and the memory includes at least one memory chip.

[0089] An embodiment of the present invention provides a computer-readable storage medium, which includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the demand processing method based on emotion perception.

[0090] Specifically, the demand processing methods based on emotion perception include:

[0091] Step S201, obtaining user interaction information, wherein the user interaction information at least includes audio information sent by the user and representing the user's needs;

[0092] Specifically, the system first needs to obtain user interaction information, which includes but is not limited to text, voice or audio information sent by users through phone calls, instant messages, emails or online forms, etc. It is particularly emphasized that audio information is a key data source for the system to recognize user emotions, because voice characteristics such as pitch, speaking speed, pauses, etc. can provide rich non-verbal emotional clues.

[0093] Step S202, determining basic user information according to the user interaction information, and analyzing the basic user information through a singular value decomposition model to obtain a user demand type, wherein the singular value decomposition model is obtained by training multiple sets of data, each set of the multiple sets of data includes historical user basic information and label data of user demand type, the user basic information includes at least a user name, and the user demand type includes any one of consultation, complaint, suggestion and report;

[0094] Specifically, after collecting the basic information of users, a user information database is constructed, and then the singular value decomposition model is trained. The system uses the historical basic information of users and their demand type labels to train through the singular value decomposition (SVD) model. SVD is a powerful matrix decomposition technology that can extract low-dimensional feature vectors from high-dimensional data, thereby simplifying the data processing process. During the training process, the model learns the intrinsic connection between the basic information of users and the demand type, establishes a user demand type prediction model, and performs real-time prediction of user demand types.

[0095] Step S203, determining the user emotion level according to the audio information in the user interaction information, and generating warning information when the user emotion level is level one to instruct the staff to handle the demand according to the user demand type, wherein the user emotion level being level one indicates that the user is in a bad mood.

[0096] Specifically, once the system determines that the user's emotional level has reached level one, an early warning message is immediately generated. The early warning message not only includes an assessment of the user's emotional state, but also includes the identification results of the user's demand type (inquiry, complaint, suggestion, report). The system sends the early warning message to the corresponding customer service or management personnel's device through SMS, email, or application notification to ensure a quick response. Staff who receive the early warning message should give priority to user requests with an emotional level of level one. During the processing process, the staff takes measures such as soothing, explaining, and responding quickly to promptly solve user problems, alleviate their negative emotions, and avoid escalation of complaints.

[0097] An embodiment of the present invention provides an electronic device, including a processor, a memory, and a program stored in the memory and executable on the processor, wherein when the processor executes the program, at least the following steps are implemented:

[0098] Step S201, obtaining user interaction information, wherein the user interaction information at least includes audio information sent by the user and representing the user's needs;

[0099] Specifically, the system first needs to obtain user interaction information, which includes but is not limited to text, voice or audio information sent by users through phone calls, instant messages, emails or online forms, etc. It is particularly emphasized that audio information is a key data source for the system to recognize user emotions, because voice characteristics such as pitch, speaking speed, pauses, etc. can provide rich non-verbal emotional clues.

[0100] Step S202, determining basic user information according to the user interaction information, and analyzing the basic user information through a singular value decomposition model to obtain a user demand type, wherein the singular value decomposition model is obtained by training multiple sets of data, each set of the multiple sets of data includes historical user basic information and label data of user demand type, the user basic information includes at least a user name, and the user demand type includes any one of consultation, complaint, suggestion and report;

[0101] Specifically, after collecting the basic information of users, a user information database is constructed, and then the singular value decomposition model is trained. The system uses the historical basic information of users and their demand type labels to train through the singular value decomposition (SVD) model. SVD is a powerful matrix decomposition technology that can extract low-dimensional feature vectors from high-dimensional data, thereby simplifying the data processing process. During the training process, the model learns the intrinsic connection between the basic information of users and the demand type, establishes a user demand type prediction model, and performs real-time prediction of user demand types.

[0102] Step S203, determining the user emotion level according to the audio information in the user interaction information, and generating warning information when the user emotion level is level one to instruct the staff to handle the demand according to the user demand type, wherein the user emotion level being level one indicates that the user is in a bad mood.

[0103] Specifically, once the system determines that the user's emotional level has reached level one, an early warning message is immediately generated. The early warning message not only includes an assessment of the user's emotional state, but also includes the identification results of the user's demand type (inquiry, complaint, suggestion, report). The system sends the early warning message to the corresponding customer service or management personnel's device through SMS, email, or application notification to ensure a quick response. Staff who receive the early warning message should give priority to user requests with an emotional level of level one. During the processing process, the staff takes measures such as soothing, explaining, and responding quickly to promptly solve user problems, alleviate their negative emotions, and avoid escalation of complaints.

[0104] The devices in this article can be servers, PCs, PADs, mobile phones, etc.

[0105] The present application also provides a computer program product, including a computer program, which, when executed by a processor, implements the steps of the method described in each embodiment of the present application:

[0106] Step S201, obtaining user interaction information, wherein the user interaction information at least includes audio information sent by the user and representing the user's needs;

[0107] Specifically, the system first needs to obtain user interaction information, which includes but is not limited to text, voice or audio information sent by users through phone calls, instant messages, emails or online forms, etc. It is particularly emphasized that audio information is a key data source for the system to recognize user emotions, because voice characteristics such as pitch, speaking speed, pauses, etc. can provide rich non-verbal emotional clues.

[0108] Step S202, determining basic user information according to the user interaction information, and analyzing the basic user information through a singular value decomposition model to obtain a user demand type, wherein the singular value decomposition model is obtained by training multiple sets of data, each set of the multiple sets of data includes historical user basic information and label data of user demand type, the user basic information includes at least a user name, and the user demand type includes any one of consultation, complaint, suggestion and report;

[0109] Specifically, after collecting the basic information of users, a user information database is constructed, and then the singular value decomposition model is trained. The system uses the historical basic information of users and their demand type labels to train through the singular value decomposition (SVD) model. SVD is a powerful matrix decomposition technology that can extract low-dimensional feature vectors from high-dimensional data, thereby simplifying the data processing process. During the training process, the model learns the intrinsic connection between the basic information of users and the demand type, establishes a user demand type prediction model, and performs real-time prediction of user demand types.

[0110] Step S203, determining the user emotion level according to the audio information in the user interaction information, and generating warning information when the user emotion level is level one to instruct the staff to handle the demand according to the user demand type, wherein the user emotion level being level one indicates that the user is in a bad mood.

[0111] Specifically, once the system determines that the user's emotional level has reached level one, an early warning message is immediately generated. The early warning message not only includes an assessment of the user's emotional state, but also includes the identification results of the user's demand type (inquiry, complaint, suggestion, report). The system sends the early warning message to the corresponding customer service or management personnel's device through SMS, email, or application notification to ensure a quick response. Staff who receive the early warning message should give priority to user requests with an emotional level of level one. During the processing process, the staff takes measures such as soothing, explaining, and responding quickly to promptly solve user problems, alleviate their negative emotions, and avoid escalation of complaints.

[0112] Obviously, those skilled in the art should understand that the above modules or steps of the present invention can be implemented by a general computing device, they can be concentrated on a single computing device, or distributed on a network composed of multiple computing devices, they can be implemented by a program code executable by a computing device, so that they can be stored in a storage device and executed by the computing device, and in some cases, the steps shown or described can be executed in a different order than here, or they can be made into individual integrated circuit modules, or multiple modules or steps therein can be made into a single integrated circuit module for implementation. Thus, the present invention is not limited to any specific combination of hardware and software.

[0113] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application may adopt the form of a computer program product implemented in one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) that include computer-usable program code.

[0114] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0115] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0116] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0117] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0118] The memory may include non-permanent memory in a computer-readable medium, random access memory (RAM) and / or non-volatile memory in the form of read-only memory (ROM) or flash RAM. The memory is an example of a computer-readable medium.

[0119] Computer readable media include permanent and non-permanent, removable and non-removable media that can be implemented by any method or technology to store information. Information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technology, compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape disk storage or other magnetic storage devices or any other non-transmission media that can be used to store information that can be accessed by a computing device. As defined herein, computer readable media does not include temporary computer readable media (transitory media), such as modulated data signals and carrier waves.

[0120] It should also be noted that the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, commodity or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, commodity or device. In the absence of more restrictions, the elements defined by the sentence "comprises a ..." do not exclude the existence of other identical elements in the process, method, commodity or device including the elements.

[0121] From the above description, it can be seen that the above embodiments of the present application achieve the following technical effects:

[0122] 1) In the demand processing method based on emotion perception of the present application, user interaction information is obtained, and the user interaction information at least includes audio information sent by the user to characterize the user demand; the user basic information is determined according to the user interaction information, and the user basic information is analyzed by a singular value decomposition model to obtain the user demand type, the singular value decomposition model is obtained by training multiple groups of data, each of the multiple groups of data includes historical user basic information and label data of the user demand type, the user basic information at least includes the user name, and the user demand type includes any one of consultation, complaint, suggestion and report; the user emotion level is determined according to the audio information in the user interaction information, and when the user emotion level is level one, an early warning message is generated to instruct the staff to process the demand according to the user demand type, and the user emotion level is level one, indicating that the user emotion is poor. Compared with the prior art, which cannot effectively identify user emotions, the present application first determines the type of user demand through a singular value decomposition model, and then determines the user emotion through audio information, and finally generates an early warning message, so that the customer demand can be processed according to the customer emotion to meet the user demand. Therefore, the problem that the power system in the prior art cannot accurately identify the user's emotions can be solved, and the effect of accurately perceiving the user's emotions can be achieved.

[0123] 2) In the demand processing device based on emotion perception of the present application, user interaction information is obtained, and the user interaction information at least includes audio information sent by the user to characterize the user demand; the user basic information is determined according to the user interaction information, and the user basic information is analyzed by a singular value decomposition model to obtain the user demand type, the singular value decomposition model is obtained by training multiple groups of data, each of the multiple groups of data includes historical user basic information and label data of the user demand type, the user basic information at least includes the user name, and the user demand type includes any one of consultation, complaint, suggestion and report; the user emotion level is determined according to the audio information in the user interaction information, and when the user emotion level is level one, an early warning message is generated to instruct the staff to process the demand according to the user demand type, and the user emotion level is level one, indicating that the user emotion is poor. Compared with the prior art, which cannot effectively identify user emotions, the present application first determines the type of user demand through a singular value decomposition model, and then determines the user emotion through audio information, and finally generates an early warning message, so that the customer demand can be processed according to the customer emotion to meet the user demand. Therefore, the problem that the power system in the prior art cannot accurately identify the user's emotions can be solved, and the effect of accurately perceiving the user's emotions can be achieved.

[0124] The above description is only the preferred 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 modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present application shall be included in the protection scope of the present application.

Claims

1. A demand processing method based on emotion perception, characterized in that: include: Acquiring user interaction information, wherein the user interaction information at least includes audio information sent by the user and representing the user's needs; Determine user basic information according to the user interaction information, and analyze the user basic information through a singular value decomposition model to obtain a user demand type, wherein the singular value decomposition model is obtained by training multiple sets of data, each set of the multiple sets of data includes historical user basic information and label data of user demand type, the user basic information includes at least a user name, and the user demand type includes any one of consultation, complaint, suggestion and report; The user emotion level is determined according to the audio information in the user interaction information. When the user emotion level is level one, an early warning message is generated to instruct the staff to handle the demand according to the user demand type, wherein the user emotion level being level one indicates that the user is in a bad mood.

2. The demand processing method based on emotion perception according to claim 1 is characterized in that: Determining the user emotion level according to the audio information in the user interaction information includes: Obtaining the decibel of the audio information, and constructing a one-to-one mapping relationship between the decibel and the corresponding time; The decibel average value is determined according to the one-to-one mapping relationship, and when the decibel average value is greater than or equal to a preset threshold, the user emotion is determined to be level one.

3. The demand processing method based on emotion perception according to claim 1 is characterized in that: Determining basic user information according to the user interaction information includes: Receiving identity authentication information sent by a user, and verifying whether the identity authentication information is consistent with pre-stored information, wherein the identity authentication information is sent by the user in response to an identity authentication request, and the pre-stored information is identity information pre-stored by the user; When the identity authentication information is consistent with the pre-stored information, the basic information of the user is acquired.

4. The demand processing method based on emotion perception according to claim 1 is characterized in that: Before analyzing the basic information of the user by using a singular value decomposition model to obtain the type of user demand, the method further includes: Acquire the historical user basic information, and perform data cleaning on the historical user basic information; The features representing the user demand types in the historical user basic information after data cleaning are labeled to obtain the label data representing the user demand types.

5. The demand processing method based on emotion perception according to claim 1 is characterized in that: The basic information of the user is analyzed by a singular value decomposition model to obtain the user demand type, including: Constructing a sparse matrix of the user's basic information and determining an initial value of the sparse matrix; An objective function is constructed according to the sparse matrix, and the objective function is solved through parameter iteration to obtain the user demand type.

6. The demand processing method based on emotion perception according to claim 1 is characterized in that: After generating the warning information, the method further includes: The warning information is sent to the staff so that the staff can make a return visit and calm the user down.

7. The demand processing method based on emotion perception according to claim 1 is characterized in that: The method further comprises: The user interaction information, the user basic information and the user demand type are stored in a preset knowledge base for use in optimizing the singular value decomposition model.

8. A demand processing device based on emotion perception, characterized in that: include: An acquisition unit, configured to acquire user interaction information, wherein the user interaction information at least includes audio information sent by the user and representing the user's needs; an analysis unit, configured to determine basic user information according to the user interaction information, and analyze the basic user information through a singular value decomposition model to obtain a user demand type, wherein the singular value decomposition model is obtained by training multiple sets of data, each set of the multiple sets of data includes historical user basic information and label data of user demand type, the user basic information includes at least a user name, and the user demand type includes any one of consultation, complaint, suggestion and report; A generating unit is used to determine the user emotion level according to the audio information in the user interaction information, and generate warning information when the user emotion level is level one to instruct the staff to handle the demand according to the user demand type, wherein the user emotion level of level one indicates that the user is in a bad mood.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein when the program is executed, the device where the computer-readable storage medium is located is controlled to execute the demand processing method based on emotion perception as described in any one of claims 1 to 7.

10. An electronic device, characterized in that: include: One or more processors, a memory, and one or more programs, wherein the one or more programs are stored in the memory and are configured to be executed by the one or more processors, and the one or more programs include a method for executing the emotion-aware demand processing method described in any one of claims 1 to 7.

Images