Customer service work order analysis method and related equipment

Abstract

The invention provides a customer service work order analysis method and related equipment. The method comprises the following steps: acquiring work order data information; analyzing the work order data information to obtain at least one analysis result; obtaining the analysis result in real time, and predicting a service quality score based on the analysis result; and performing early warning monitoring on the service quality based on the service quality score. According to the embodiment of the invention, the multi-dimensional information in the customer service work order is comprehensively and deeply mined, the relevance and the deep-level value among the dimensions are revealed, an enterprise can more accurately know the actual condition and the service quality of customer service work, and comprehensive and reliable data support is provided for decision making.

Description

Technical Field

[0001] The present application relates to the technical field of work order analysis, and in particular to a customer service work order analysis method and related equipment. Background Art

[0002] In the current customer service management field, customer service ticket analysis is an important means to evaluate service quality, optimize service processes, and improve customer satisfaction. The current customer service ticket analysis method can often only analyze customer service tickets in a single dimension, such as focusing only on surface indicators such as the number of tickets and processing time, while ignoring the relevance and deep-level value of multi-dimensional information such as customers, products, and service processes. The single-dimensional analysis method is difficult to fully and accurately reflect the actual situation and service quality of customer service work, resulting in one-sided analysis results and insufficient decision-making basis.

[0003] Moreover, the current analysis methods lack intelligent early warning mechanisms. When service quality has problems or declines, it is often impossible to detect and intervene in time, resulting in the accumulation of problems and ultimately affecting customer satisfaction and loyalty. This passive service model cannot meet the needs of modern customers for efficient and personalized services. Summary of the invention

[0004] In view of this, the purpose of this application is to propose a customer service work order analysis method and related equipment.

[0005] Based on the above purpose, this application provides a customer service work order analysis method, including:

[0006] Get work order data information;

[0007] Analyze the work order data information to obtain at least one analysis result;

[0008] Acquire the analysis results in real time, and predict a service quality score based on the analysis results;

[0009] The service quality is monitored based on the service quality score.

[0010] In a possible implementation, the work order data information includes work order information; the analysis result includes a first result and a second result;

[0011] The analyzing the work order data information to obtain at least one analysis result includes:

[0012] Analyze the work order receiving volume, work order processing volume and processing time of each month of the work order information to determine the peak period and the trough period of work order processing, and obtain the first result;

[0013] The second result is obtained by analyzing data changes of the work order information in different seasons.

[0014] In a possible implementation, the work order data information includes customer information; the analysis result includes a third result and a fourth result;

[0015] The analyzing the work order data information to obtain at least one analysis result includes:

[0016] Based on the customer type in the customer information, analyzing the work order processing efficiency and customer satisfaction index of the customer information to obtain the third result;

[0017] Based on the regions where the customers in the customer information are distributed, the work order submission volume and processing time of the customer information in different regions are analyzed to obtain the fourth result.

[0018] In a possible implementation, the work order data information includes product information; the analysis result includes a fifth result and a sixth result;

[0019] The analyzing the work order data information to obtain at least one analysis result includes:

[0020] Analyze the proportion of fault work orders and fault causes in the product information to obtain the fifth result;

[0021] The sixth result is obtained by analyzing the number of product work orders and the difficulty of processing the product information at different life cycle stages.

[0022] In a possible implementation, acquiring the analysis result in real time and predicting a service quality score based on the analysis result includes:

[0023] At least one of the analysis results is input into a pre-trained service quality warning model. In the service quality warning model, based on the analysis results, the weight corresponding to each analysis result is determined using trained parameters for prediction to obtain the service quality score.

[0024] In a possible implementation, after obtaining the work order data information, the method further includes:

[0025] The missing values ​​and abnormal values ​​in the work order data information are processed, and the work order data information is standardized.

[0026] Based on the same inventive concept, the embodiment of the present application also provides a customer service work order analysis device, including:

[0027] An acquisition module is configured to acquire work order data information;

[0028] An analysis module, configured to analyze the work order data information to obtain at least one analysis result;

[0029] A prediction module is configured to obtain the analysis result in real time and predict a service quality score based on the analysis result;

[0030] The monitoring module is configured to perform early warning monitoring on the service quality based on the service quality score.

[0031] Based on the same inventive concept, an embodiment of the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, a customer service work order analysis method as described in any one of the above items is implemented.

[0032] Based on the same inventive concept, an embodiment of the present application also provides a non-transitory computer-readable storage medium, which stores computer instructions, and the computer instructions are used to enable the computer to execute any of the above-mentioned customer service work order analysis methods.

[0033] Based on the same inventive concept, an embodiment of the present application also provides a computer program product, which includes computer program instructions, and the computer instructions are used to enable the computer program product to execute any of the customer service work order analysis methods described above.

[0034] From the above, it can be seen that the customer service work order analysis method and related equipment provided by the present application obtain work order data information; analyze the work order data information to obtain at least one analysis result; obtain the analysis result in real time, and predict the service quality score based on the analysis result; and perform early warning monitoring on the service quality based on the service quality score. The embodiment of the present application reveals the correlation and deep-level value between the dimensions by comprehensively and deeply mining the multi-dimensional information in the customer service work order, which helps enterprises to more accurately understand the actual situation and service quality of customer service work and provide comprehensive and reliable data support for decision-making. In addition, the present application can continuously monitor the output results of the analysis models of each dimension, predict the service quality score, and promptly discover and trigger the early warning mechanism when there are problems or declines in service quality, which helps enterprises to take timely measures to intervene and prevent problems from accumulating, thereby improving customer satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the technical solutions in the present application or related technologies, the drawings required for use in the embodiments or related technical descriptions are briefly introduced below. Obviously, the drawings described below are only embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0036] Figure 1 A flowchart of a customer service work order analysis method according to an embodiment of the present application;

[0037] Figure 2 This is a schematic diagram of a customer service work order analysis device according to an embodiment of the present application;

[0038] Figure 3 This is a schematic diagram of the structure of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0039] In order to make the objectives, technical solutions and advantages of the present application more clearly understood, the present application is further described in detail below in combination with specific embodiments and with reference to the accompanying drawings.

[0040] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the embodiments of the present application should be the usual meanings understood by people with ordinary skills in the field to which the present application belongs. The "first", "second" and similar words used in the embodiments of the present application do not represent any order, quantity or importance, but are only used to distinguish different components. "Including" or "comprising" and similar words mean that the elements or objects appearing in front of the word cover the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0041] It is understandable that before using the technical solutions of each embodiment of the present disclosure, the type, scope of use, usage scenarios, etc. of the personal information involved will be informed to the user in an appropriate manner, and the user's authorization will be obtained.

[0042] For example, in response to receiving an active request from a user, a prompt message is sent to the user to clearly remind the user that the operation requested to be performed will require obtaining and using the user's personal information. Thus, the user can independently choose whether to provide personal information to software or hardware such as an electronic device, application, server, or storage medium that performs the operation of the technical solution of the present disclosure according to the prompt message.

[0043] As an optional but non-limiting implementation, in response to receiving the user's active request, the prompt information may be sent to the user in the form of a pop-up window, in which the prompt information may be presented in text form. In addition, the pop-up window may also carry a selection control for the user to choose "agree" or "disagree" to provide personal information to the electronic device.

[0044] It is understandable that the above notification and the process of obtaining user authorization are merely illustrative and do not constitute a limitation on the implementation of the present disclosure. Other methods that meet relevant laws and regulations may also be applied to the implementation of the present disclosure.

[0045] As mentioned in the background technology section, in the current customer service management field, customer service ticket analysis is an important means to evaluate service quality, optimize service processes, and improve customer satisfaction. The current customer service ticket analysis method can often only analyze customer service tickets in a single dimension, such as focusing only on surface indicators such as the number of tickets and processing time, while ignoring the relevance and deep-level value of multi-dimensional information such as customers, products, and service processes. The single-dimensional analysis method is difficult to fully and accurately reflect the actual situation and service quality of customer service work, resulting in one-sided analysis results and insufficient decision-making basis.

[0046] Moreover, the current analysis methods lack intelligent early warning mechanisms. When service quality has problems or declines, it is often impossible to detect and intervene in time, resulting in the accumulation of problems and ultimately affecting customer satisfaction and loyalty. This passive service model cannot meet the needs of modern customers for efficient and personalized services.

[0047] Taking the above into consideration, the embodiment of the present application proposes a customer service work order analysis method, which obtains work order data information; analyzes the work order data information to obtain at least one analysis result; obtains the analysis result in real time, and predicts the service quality score based on the analysis result; and performs early warning monitoring of the service quality based on the service quality score. The embodiment of the present application reveals the correlation and deep-seated value between the dimensions by comprehensively and deeply mining the multi-dimensional information in the customer service work order, which helps enterprises to more accurately understand the actual situation and service quality of customer service work, and provide comprehensive and reliable data support for decision-making. In addition, the present application can continuously monitor the output results of the analysis models of each dimension, predict the service quality score, and promptly detect and trigger the early warning mechanism when there is a problem or decline in service quality, which helps enterprises to take timely measures to intervene and prevent problems from accumulating, thereby improving customer satisfaction.

[0048] The technical solutions of the embodiments of the present application are described in detail below through specific examples.

[0049] refer to Figure 1 The customer service work order analysis method of the embodiment of the present application includes the following steps:

[0050] Step S101, obtaining work order data information;

[0051] Step S102, analyzing the work order data information to obtain at least one analysis result;

[0052] Step S103, obtaining the analysis result in real time, and predicting a service quality score based on the analysis result;

[0053] Step S104: performing early warning monitoring on the service quality based on the service quality score.

[0054] With respect to step S101, original data related to customer service work orders is accessed regularly, including work order information, customer information, and product information.

[0055] In this embodiment, original data related to customer service work orders are regularly accessed from the customer service system, including but not limited to work order information (such as work order number, creation time, processing time, processing status, etc.), customer information (such as customer ID, customer type, geographical distribution, historical interaction records, etc.) and product information (such as product ID, product type, fault records, life cycle stage, etc.).

[0056] Furthermore, in some embodiments, after obtaining the work order data information, the method further includes: processing missing values ​​and abnormal values ​​in the work order data information, and standardizing the work order data information.

[0057] In this embodiment, the raw data is cleaned to remove duplicate, invalid and abnormal data records. At the same time, the data is standardized, including the unification of data format, normalization or standardization of data values, etc., to form a regular data set.

[0058] Specifically, the method of data standardization is: using the Z-score standardization method, subtracting the mean of each numerical data and dividing it by its standard deviation, so that the processed data conforms to the standard normal distribution. The formula for standardization is: Among them, X is the original data, μ is the mean of the original data, σ is the standard deviation of the original data, and Z is the standardized data.

[0059] Data cleaning methods include:

[0060] Missing value processing, using linear interpolation algorithm to fill;

[0061] For outlier processing, the threshold judgment method is used to treat data points that exceed the preset threshold range as outliers and replace them with the mean of their neighboring points.

[0062] For data cleaning, this embodiment mainly includes two methods: missing value processing and outlier processing. For missing value processing, a linear interpolation algorithm is used for filling. Specifically, when a missing value is detected in a data set, the missing value is filled using the linear interpolation result of the two valid values ​​before and after the missing value. For example, assuming that the data sequence is [1,2,? ,5,6], where "?" represents a missing value, the filling value calculated by the linear interpolation algorithm is 3 (i.e. (2+5) / 2=3.5, rounded to 3), so the filled data sequence becomes [1,2,3,5,6].

[0063] For outlier processing, the threshold judgment method is used. Specifically, first set a reasonable threshold range based on the data distribution and business needs. Then, when a data point is detected to be beyond this preset threshold range, it is regarded as an outlier and replaced with the mean of its neighboring points. For example, suppose a threshold range of [1,5] is set for a feature "satisfaction score". When a data point 7 is detected, it is regarded as an outlier and replaced with the mean of its neighboring points (assumed to be 5 and another valid value 4) of 4.5 (rounded to 5 if an integer is required).

[0064] Through the above data standardization and data cleaning steps, a cleaner and more standardized data set can be obtained, providing a data foundation for subsequent data modeling and model training.

[0065] With respect to step S102, the work order data information is analyzed to obtain at least one analysis result.

[0066] In some embodiments, the work order data information includes work order information; the analysis result includes a first result and a second result; the work order data information is analyzed to obtain at least one analysis result, including: analyzing the work order reception volume, work order processing volume and processing time of each month of the work order information to determine the peak and trough periods of work order processing to obtain the first result; analyzing the data changes of the work order information in different seasons to obtain the second result.

[0067] In some embodiments, the work order data information includes customer information; the analysis results include a third result and a fourth result; the work order data information is analyzed to obtain at least one analysis result, including: based on the customer type in the customer information, analyzing the work order processing efficiency and customer satisfaction index of the customer information to obtain the third result; based on the regions where the customers are distributed in the customer information, analyzing the work order submission volume and processing time of the customer information in different regions to obtain the fourth result.

[0068] In some embodiments, the work order data information includes product information; the analysis results include a fifth result and a sixth result; the work order data information is analyzed to obtain at least one analysis result, including: analyzing the proportion of faulty work orders and the causes of faults in the product information to obtain the fifth result; analyzing the number and processing difficulty of product work orders at different life cycle stages of the product information to obtain the sixth result.

[0069] In the embodiments of the present application, a comprehensive and in-depth analysis of customer service work order data is conducted by constructing data analysis models in multiple dimensions, with the aim of improving the quality of customer service, improving service efficiency, and enhancing overall customer satisfaction. These analysis models cover multiple important dimensions, including time dimension, customer dimension, product dimension, and service dimension. Each dimension is approached from a different perspective to fully understand the dynamic changes of work order data, thereby optimizing service processes and decisions.

[0070] In the time dimension analysis model, a monthly work order trend analysis unit and a seasonal fluctuation analysis unit are set up. The combination of these two units can help form a deep understanding of time changes. The monthly work order trend analysis unit can clearly identify the peak and trough periods of work order processing by tracking the number of work orders received, processed, and processed each month. By reviewing and analyzing historical data, it can be found that the work order fluctuations caused by specific months, such as around holidays, the number of work orders received tends to increase significantly. The identification of this trend not only provides a basis for the reasonable allocation of human resources, but also helps to prepare in advance to ensure that the customer service team can respond efficiently when customer demand surges. In addition, the seasonal fluctuation analysis unit focuses on products that are greatly affected by seasonality and analyzes the changes in the types and quantity of work orders in different seasons. For example, for seasonal products such as air conditioners and heaters, it can be observed that the number of related fault work orders will rise sharply in summer and winter. Through this analysis, it can not only provide support for product promotion and marketing strategies, but also take into account seasonal needs in advance in service design and make reasonable allocation of resources.

[0071] For example, in the time dimension analysis model, the monthly work order trend analysis unit can analyze the customer service work order data of an e-commerce platform and find that the number of work orders received during the Double Eleven Shopping Festival every year will surge. For example, during the Double Eleven period of a certain year, the number of work orders received increased by 150% compared to usual. The analysis results show that they are mainly concentrated in areas such as order processing, delivery delays, and payment issues. With such data, companies can make resource allocations in advance and increase customer service staff to ensure timely response in the case of a surge in customer demand and reduce customer waiting time. At the same time, the seasonal fluctuation analysis unit can analyze the fault work order data of air-conditioning products and find that the number of work orders in summer has increased significantly, mainly due to refrigeration failures and noise problems. Through this analysis, companies can prepare in advance, increase maintenance and support for related products, and ensure that customers enjoy high-quality services.

[0072] The customer dimension analysis model starts from the customer's perspective and is subdivided into customer type analysis unit and regional distribution analysis unit. The customer type analysis unit classifies customer groups and analyzes the performance of different types of customers in the work order processing process, including work order processing efficiency and customer satisfaction indicators. For example, VIP customers usually expect higher service quality, while new customers may not be familiar with product and service processes, so customized service solutions need to be provided for different types of customers. The regional distribution analysis unit takes into account geographical differences and identifies the impact of regional differences on customer service efficiency by analyzing the number of work order submissions and processing time of customers in different regions. This analysis not only helps understand which regions may face more problems for customers, but also provides data support for the formulation of regional service strategies to ensure that they can be optimized for the characteristics of specific regions.

[0073] For example, the customer service department of a software company analyzed the performance of VIP customers and ordinary customers in the work order processing process and found that the average processing time for VIP customers' work orders was 30 minutes, while that for ordinary customers was 50 minutes. Through further analysis, it was found that the work order content of VIP customers was more complicated, but due to their importance, the customer service team provided higher priority support to VIP customers. In order to improve the satisfaction of ordinary customers, the company decided to provide more detailed self-service documents for ordinary customers to reduce the frequency of their inquiries. At the same time, the application of the regional distribution analysis unit can reveal that customers in a certain area have more feedback on the product. Assume that in a certain city in the south, the number of work orders submitted by customers is significantly higher than the national average. After analysis, it is found that the climate conditions in the region have led to an increase in the frequency of product failures. Based on this data, the company can consider adding technical support personnel in the region and launch customized service solutions based on the climate characteristics of the region.

[0074] The product dimension analysis model includes the product failure rate analysis unit and the product life cycle analysis unit. The product failure rate analysis unit focuses on the in-depth analysis of the proportion of failure orders for specific products and their causes. By counting and classifying the failure data, it is possible to identify which products frequently have problems during use, and conduct in-depth discussions on the causes of failures, thereby providing a scientific basis for product improvement and quality control. Through this analysis, not only can the reliability of the product be improved, but also the problem can be quickly located when a failure occurs, providing customers with more efficient solutions. In addition, the product life cycle analysis unit focuses on the number of work orders and the difficulty of handling the product at different stages of its life cycle. By analyzing the characteristics of work orders in the introduction period, growth period, maturity period and decline period, it can provide a basis for the company to formulate corresponding market strategies at various stages of the product. This analysis can help identify the challenges that may be encountered at different stages, so as to formulate effective countermeasures.

[0075] For example, suppose that through data analysis, it is found that the proportion of faulty work orders within the first six months of use of this mobile phone is as high as 12%, and the causes of failure are mainly concentrated in battery and display problems. Through in-depth analysis of these data, companies can quickly adjust production lines, improve product design, and even launch free battery replacement services or display protection measures for existing customers to improve customer satisfaction. The product life cycle analysis unit can analyze the performance of a newly launched smart watch in the market. Suppose that the number of work orders is small during the introduction period, but in the growth period, the number of work orders gradually increases. Through analysis, it is found that the work orders of the product in the growth period are mainly concentrated in software updates and function usage issues. This data suggests that companies need to strengthen product training and software support at this stage to ensure that customers can fully utilize the various functions of the product.

[0076] The service dimension analysis model starts from the perspective of service, and mainly includes the service type analysis unit and the response time analysis unit. The service type analysis unit classifies work orders by service type to obtain the processing efficiency and customer satisfaction of various services. This analysis can not only help identify which service categories perform well and which need improvement, but also provide a basis for service training and optimization, so that the customer service team can better meet customer needs. The response time analysis unit focuses on analyzing the response time and processing time of work orders, and identifying the key factors that affect processing efficiency. By monitoring and analyzing the response time, the root cause of the response delay can be found, and corresponding measures can be taken to improve it, thereby improving the overall service efficiency.

[0077] For example, suppose that a courier company's customer service system analyzes work orders of different service types and finds that the processing efficiency of intra-city express delivery is as high as 90%, while the processing efficiency of international express delivery is only 60%. After analysis, it is found that the customs clearance and transportation time issues involved in the processing of international express work orders lead to longer processing time. In order to improve the service quality of international express delivery, the company decides to communicate with relevant logistics partners to improve customs clearance efficiency and increase the customer service team's expertise in this field to speed up processing. In addition, the application of the response time analysis unit can reveal the key factors affecting processing efficiency. Suppose through data analysis, it is found that the average response time for a certain type of work order is 10 minutes, while other types of work orders are only 5 minutes. After in-depth analysis, it is found that the main reason for the delay is the lack of relevant knowledge base support. Therefore, the company decides to establish a knowledge base for this type of work order to help customer service find solutions faster, thereby improving response speed.

[0078] Combining the analysis of the above dimensions, we can build a comprehensive and three-dimensional customer service work order data analysis system, which can not only monitor and evaluate all aspects of customer service in real time, but also provide a scientific basis for decision-making, helping enterprises stand out in the increasingly fierce market competition. Through data-driven decision-making, more efficient resource allocation and more accurate customer service can be achieved, thereby improving customer satisfaction and loyalty, and ultimately achieving sustainable development of the enterprise.

[0079] In some embodiments, the monthly work order trend analysis unit uses a moving average time series analysis algorithm to perform trend analysis on the monthly work order receipt volume, processing volume, and processing time, that is, the first result is calculated by the following formula:

[0080]

[0081] Among them, F t is the predicted value for month t, x i is the actual value of the ith month, and n is the period of the moving average.

[0082] In this embodiment, monthly work order receipt volume, processing volume, and processing time data are extracted from the customer service system to form a time series data set.

[0083] Select an appropriate moving average period n. The selection of period n should be based on the characteristics of the data and the analysis requirements. Usually, you can choose 3 months, 6 months or 12 months, etc.

[0084] For each monthly data point, the moving average of the previous n months is calculated as the predicted value for that month.

[0085] Based on the calculated moving average, draw a monthly trend chart of the number of work orders received, the number of work orders processed, and the processing time. Analyze the long-term trends and cyclical changes in the trend chart to identify the peak and trough periods of work order processing, as well as possible seasonal fluctuations.

[0086] Assuming that 6 months is selected as the moving average period, for the forecast of the number of work orders received in June 2023, the average value of the actual number of receipts from January 2023 to June 2023 will be calculated as the forecast value. Similarly, the same moving average algorithm is used to calculate the forecast of processing volume and processing time.

[0087] By analyzing the trend chart, we can find that the number of work orders received in certain months (such as around holidays) increases significantly, and the processing time may be extended accordingly, so as to adjust service resources in time and optimize service processes.

[0088] The results of monthly work order trend analysis are used as an important basis for evaluating service quality, formulating service plans and optimizing service processes. Based on the results of trend analysis, service resources are rationally allocated to improve service efficiency, ensure that customer work orders can be processed in a timely manner during peak periods, and improve customer satisfaction.

[0089] With respect to step S103, the analysis result is acquired in real time, and a service quality score is predicted based on the analysis result.

[0090] In some embodiments, the real-time acquisition of the analysis results and the prediction of the service quality score based on the analysis results include: inputting at least one of the analysis results into a pre-trained service quality warning model, in which the weight corresponding to each analysis result is determined based on the analysis results using trained parameters for prediction to obtain the service quality score.

[0091] In this embodiment, a linear regression algorithm is used to construct a service quality early warning model, and the specific steps are as follows:

[0092] Collect the result data output by the analysis models of each dimension during the customer service ticket analysis process as the input data of the service quality early warning model, including the analysis results of the time dimension, customer dimension, product dimension and service dimension.

[0093] Through data standardization, the impact of dimensional differences on model training is eliminated to form a regular data set; the processed data set is divided into training set, validation set and test set, where the training set is used for model training, the validation set is used to verify and optimize the model effect, and the test set is used to verify the generalization ability of the model on unknown data sets.

[0094] The linear regression model is selected as the service quality early warning model, which can predict the service quality score based on the analysis results of each dimension of the input.

[0095] Specifically, in this embodiment, we use a linear regression algorithm to build a service quality early warning model, and the specific steps are as follows. First, we collect the result data output by the analysis model of each dimension during the customer service work order analysis process, and these data will be used as the input of the service quality early warning model. The collected data includes the number of work orders received and the processing time in the time dimension, the proportion of customer types in the customer dimension, the failure rate in the product dimension, and the response time and customer satisfaction in the service dimension. These data will provide us with a comprehensive perspective to understand the various factors that affect service quality.

[0096] Next, in order to eliminate the impact of dimensional differences on model training, we standardized the collected data. This process converts all input data into a relatively uniform dimension, allowing the model to learn the relationship between features more effectively. After standardization, we formed a regular data set. In order to improve the training effect of the model, we divided the processed data set into three parts: training set, validation set, and test set. The training set is used for model training, the validation set is used to verify and optimize the model effect, and the test set is used to verify the generalization ability of the model on unknown data sets, thereby ensuring that the model can have good prediction performance in practical applications.

[0097] When choosing a model, we decided to use a linear regression model as a service quality early warning model. Linear regression is an effective tool that can predict service quality scores based on the analysis results of various input dimensions. For example, suppose we analyze the customer service ticket data for the past three months and find that there is a certain linear relationship between the increase in the number of tickets received and the decrease in customer satisfaction. Through the linear regression model, we can establish an equation to predict the change in service quality scores under certain conditions.

[0098] When adjusting model parameters, we use the validation set to test the impact of different parameter settings on model performance. Through cross-validation, we can find the optimal parameter combination to improve the accuracy and stability of the model. After training, we apply the model to the test set to evaluate its performance on unknown data. Suppose in actual application, our model successfully predicts the service quality score within a certain period of time and identifies potential risk factors, such as customer complaints caused by long response time or high failure rate products.

[0099] For example, in an actual application, our model predicted that the service quality score would drop significantly in a certain week, mainly because the failure rate of a newly released product in the market increased abnormally. Through the analysis of the linear regression model, we can issue early warnings in time and recommend relevant departments to increase support for the product, such as sending more technicians to provide on-site services and providing more detailed usage guidance, thereby reducing customer complaints and dissatisfaction caused by product problems.

[0100] Ultimately, through this linear regression service quality early warning model, enterprises can issue early warnings before service quality declines, take measures to intervene in advance, and ensure the maintenance of customer satisfaction. This early warning mechanism based on data analysis not only improves the enthusiasm and responsiveness of customer service work, but also effectively reduces customer churn caused by service quality issues and enhances the market competitiveness of enterprises. By continuously iterating and optimizing the service quality early warning model, enterprises will be able to continuously improve customer experience and service quality in a fierce market environment.

[0101] The algorithm formula of the linear regression model is:

[0102] y=β0+β1x1+β2x2+…+β n x n +ε

[0103] Among them, y is the target variable, representing the predicted service quality score, x1, x2, ..., x n is the input variable, representing the result data output by the analysis model of each dimension; β1, β2, ..., β n is the regression coefficient, which is determined by model training; β0 is the intercept term; ε is the error term; the purpose of model training is to estimate the regression coefficients β0, β1, β2, ..., β n , so that the error between the predicted value and the true value is minimized.

[0104] Furthermore, for each feature x in the linear regression model i The corresponding parameter β i Assign an initial value, which is set to a random number close to 0, and also assign an initial value to the intercept term β0; these initial parameter values ​​will serve as the starting point for model training and will be continuously optimized and adjusted through subsequent training processes.

[0105] The training set is then used to train the linear regression model, and the trained and verified linear regression model is saved as a service quality early warning model for subsequent continuous monitoring of the output results of the analysis model in each dimension and prediction of the service quality score.

[0106] The training method of the service quality prediction model is: adopt the gradient descent training algorithm to iteratively train the model; in each round of iteration, calculate the difference between the prediction result of the model under the current parameters and the true label, that is, the loss function value, and update the model parameters through the back propagation algorithm to minimize the loss function; in the gradient descent algorithm, the parameter update formula is expressed as:

[0107]

[0108] Among them, θ represents the model parameters, including all weights and biases that the model needs to learn; α represents the learning rate, which is used to control the step size of parameter update; It represents the gradient of the loss function J(θ) with respect to the parameter θ, which is a vector pointing to the direction in which the loss function grows fastest; := represents the assignment operation, that is, updating the value of the parameter θ.

[0109] The above loss function may be a mean square error.

[0110] Furthermore, the trained model is verified using the validation set, and the model parameters are adjusted based on the verification results; the optimized model is comprehensively evaluated using the test set using the set evaluation indicators to ensure that the model performance meets the expected standards.

[0111] In another feasible embodiment, it is demonstrated how to use a linear regression model and a gradient descent algorithm to build and train a service quality prediction model. Specifically, it includes:

[0112] Data preparation: Use the numpy library to generate some random data to simulate customer service ticket data. X represents the feature matrix, each row represents a ticket, and each column represents a feature; y represents the target variable, that is, the service quality score. Use the train_test_split function to divide the dataset into training set, validation set, and test set.

[0113] Model initialization: An initialize_parameters function is defined to initialize the model parameters, including weights W and biases b. In this example, the weights are initialized to the zero vector and the biases are initialized to zero.

[0114] Model definition: A linear_model function is defined to implement the linear regression model, that is, to calculate the dot product of the input feature X and the weight W, and add the bias b.

[0115] Loss function: A compute_loss function is defined to calculate the difference between the model prediction result and the true label, that is, the mean square error (MSE).

[0116] Gradient calculation: A compute_gradients function is defined to calculate the gradient of the loss function with respect to the weight W and bias b. These gradients will be used to update the model parameters.

[0117] Gradient descent algorithm: A gradient_descent function is defined to implement the gradient descent algorithm. This function accepts input feature X, true label y, weight W, bias b, learning rate and number of iterations as parameters, and returns the trained weight and bias.

[0118] In each iteration, the function first calculates the model predictions, loss values, and gradients, and then updates the weights and biases using the gradient descent update formula.

[0119] Model training and validation: Set hyperparameters such as learning rate and number of iterations, call the initialize_parameters function to initialize the parameters, call the gradient_descent function to train the model, and get the trained weights and biases, use the trained model to make predictions on the validation set, and calculate the validation loss.

[0120] It can be seen from the above embodiments that the customer service work order analysis method described in the embodiments of the present application obtains work order data information; analyzes the work order data information to obtain at least one analysis result; obtains the analysis result in real time, and predicts the service quality score based on the analysis result; and performs early warning monitoring of the service quality based on the service quality score. The embodiments of the present application reveal the correlation and deep-level value between the dimensions by comprehensively and deeply mining the multi-dimensional information in the customer service work order, which helps enterprises to more accurately understand the actual situation and service quality of customer service work and provide comprehensive and reliable data support for decision-making. In addition, the present application can continuously monitor the output results of the analysis models of each dimension, predict the service quality score, and promptly discover and trigger the early warning mechanism when there are problems or declines in service quality, which helps enterprises to take timely measures to intervene and prevent problems from accumulating, thereby improving customer satisfaction.

[0121] It should be noted that the method of the embodiment of the present application can be performed by a single device, such as a computer or server. The method of this embodiment can also be applied to a distributed scenario and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only perform one or more steps in the method of the embodiment of the present application, and the multiple devices will interact with each other to complete the described method.

[0122] It should be noted that the above describes some embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recorded in the claims can be performed in an order different from that in the above embodiments and still achieve the desired results. In addition, the processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0123] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a customer service work order analysis device.

[0124] refer to Figure 2 , the customer service work order analysis device comprises:

[0125] The acquisition module 21 is configured to acquire work order data information;

[0126] An analysis module 22 is configured to analyze the work order data information to obtain at least one analysis result;

[0127] The prediction module 23 is configured to obtain the analysis result in real time and predict the service quality score based on the analysis result;

[0128] The monitoring module 24 is configured to perform early warning monitoring on the service quality based on the service quality score.

[0129] For the convenience of description, the above device is described in terms of functions divided into various modules. Of course, when implementing the present application, the functions of each module can be implemented in the same or multiple software and / or hardware.

[0130] The device of the above embodiment is used to implement the corresponding customer service work order analysis method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0131] Based on the same inventive concept, corresponding to any of the above-mentioned embodiments, the present application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the customer service work order analysis method described in any of the above embodiments is implemented.

[0132] Figure 3A more specific schematic diagram of the hardware structure of an electronic device provided in this embodiment is shown, and the device may include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 in the device.

[0133] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an application-specific integrated circuit (ASIC), or one or more integrated circuits, and is used to execute relevant programs to implement the technical solutions provided in the embodiments of this specification.

[0134] The memory 1020 may be implemented in the form of ROM (Read Only Memory), RAM (Random Access Memory), static storage device, dynamic storage device, etc. The memory 1020 may store an operating system and other application programs. When the technical solutions provided in the embodiments of this specification are implemented by software or firmware, the relevant program codes are stored in the memory 1020 and are called and executed by the processor 1010.

[0135] The input / output interface 1030 is used to connect the input / output module to realize information input and output. The input / output module can be configured in the device as a component (not shown in the figure), or it can be externally connected to the device to provide corresponding functions. The input device may include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device may include a display, a speaker, a vibrator, an indicator light, etc.

[0136] The communication interface 1040 is used to connect a communication module (not shown) to realize communication interaction between the device and other devices. The communication module can realize communication through a wired mode (such as USB, network cable, etc.) or a wireless mode (such as mobile network, WIFI, Bluetooth, etc.).

[0137] The bus 1050 includes a path that transmits information between the various components of the device (eg, the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040).

[0138] It should be noted that, although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device may also include other components necessary for normal operation. In addition, it can be understood by those skilled in the art that the above device may also only include the components necessary for implementing the embodiments of the present specification, and does not necessarily include all the components shown in the figure.

[0139] The electronic device of the above embodiment is used to implement the corresponding customer service work order analysis method in any of the above embodiments, and has the beneficial effects of the corresponding method embodiment, which will not be repeated here.

[0140] Based on the same inventive concept, corresponding to any of the above-mentioned embodiment methods, the present application also provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores computer instructions, and the computer instructions are used to enable the computer to execute the customer service work order analysis method described in any of the above embodiments.

[0141] The computer-readable medium of this embodiment includes permanent and non-permanent, removable and non-removable media, and information storage can be implemented by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, 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, read-only compact disk read-only memory (CD-ROM), digital versatile disk (DVD) or other optical storage, magnetic cassettes, magnetic tape magnetic 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.

[0142] The computer instructions stored in the storage medium of the above embodiment are used to enable the computer to execute the customer service work order analysis method described in any of the above embodiments, and have the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0143] Based on the same inventive concept, corresponding to the customer service work order analysis method described in any of the above embodiments, the present disclosure also provides a computer program product, which includes computer program instructions. In some embodiments, the computer program instructions can be executed by one or more processors of a computer so that the computer and / or the processor executes the customer service work order analysis method. Corresponding to the execution subject corresponding to each step in each embodiment of the customer service work order analysis method, the processor that executes the corresponding step can belong to the corresponding execution subject.

[0144] The computer program product of the above embodiment is used to enable the computer and / or the processor to execute the customer service work order analysis method described in any of the above embodiments, and has the beneficial effects of the corresponding method embodiments, which will not be repeated here.

[0145] Those skilled in the art should understand that the discussion of any of the above embodiments is merely illustrative and is not intended to imply that the scope of the present application (including the claims) is limited to these examples. In line with the concept of the present application, the technical features in the above embodiments or different embodiments may be combined, the steps may be implemented in any order, and there are many other variations of the different aspects of the embodiments of the present application as described above, which are not provided in detail for the sake of simplicity.

[0146] In addition, to simplify the description and discussion, and in order not to make the embodiments of the present application difficult to understand, the known power supply / ground connection with the integrated circuit (IC) chip and other components may or may not be shown in the provided drawings. In addition, the device can be shown in the form of a block diagram to avoid making the embodiments of the present application difficult to understand, and this also takes into account the fact that the details of the implementation of these block diagram devices are highly dependent on the platform to be implemented in the embodiments of the present application (that is, these details should be fully within the scope of understanding of those skilled in the art). In the case of elaborating specific details (e.g., circuits) to describe exemplary embodiments of the present application, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details or when these specific details are changed. Therefore, these descriptions should be considered to be illustrative rather than restrictive.

[0147] Although the present application has been described in conjunction with specific embodiments of the present application, many replacements, modifications and variations of these embodiments will be apparent to those skilled in the art from the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) may use the embodiments discussed.

[0148] The embodiments of the present application are intended to cover all such substitutions, modifications and variations that fall within the broad scope of the appended claims. Therefore, any omissions, modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the embodiments of the present application should be included in the scope of protection of the present application.

Claims

1. A customer service work order analysis method, characterized in that: include: Get work order data information; Analyze the work order data information to obtain at least one analysis result; Acquire the analysis results in real time, and predict a service quality score based on the analysis results; The service quality is monitored based on the service quality score.

2. The method according to claim 1, characterized in that The work order data information includes work order information; the analysis result includes a first result and a second result; The analyzing the work order data information to obtain at least one analysis result includes: Analyze the work order receiving volume, work order processing volume and processing time of each month of the work order information to determine the peak period and the trough period of work order processing, and obtain the first result; The second result is obtained by analyzing data changes of the work order information in different seasons.

3. The method according to claim 1, characterized in that The work order data information includes customer information; The analysis results include a third result and a fourth result; The analyzing the work order data information to obtain at least one analysis result includes: Based on the customer type in the customer information, analyzing the work order processing efficiency and customer satisfaction index of the customer information to obtain the third result; Based on the regions where the customers in the customer information are distributed, the work order submission volume and processing time of the customer information in different regions are analyzed to obtain the fourth result.

4. The method according to claim 1, characterized in that: The work order data information includes product information; the analysis result includes a fifth result and a sixth result; The analyzing the work order data information to obtain at least one analysis result includes: Analyze the proportion of fault work orders and fault causes in the product information to obtain the fifth result; The sixth result is obtained by analyzing the number of product work orders and the difficulty of processing the product information at different life cycle stages.

5. The method according to claim 1, characterized in that The real-time acquisition of the analysis result and prediction of the service quality score based on the analysis result include: At least one of the analysis results is input into a pre-trained service quality warning model. In the service quality warning model, based on the analysis results, the weight corresponding to each analysis result is determined using trained parameters for prediction to obtain the service quality score.

6. The method according to claim 1, characterized in that After obtaining the work order data information, the method further includes: The missing values ​​and abnormal values ​​in the work order data information are processed, and the work order data information is standardized.

7. A customer service work order analysis device, characterized in that: include: An acquisition module is configured to acquire work order data information; An analysis module, configured to analyze the work order data information to obtain at least one analysis result; A prediction module is configured to obtain the analysis result in real time and predict a service quality score based on the analysis result; The monitoring module is configured to perform early warning monitoring on the service quality based on the service quality score.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the program, the method according to any one of claims 1 to 6 is implemented.

9. A non-transitory computer-readable storage medium storing computer instructions, characterized in that: The computer instructions are used to enable a computer to execute the method according to any one of claims 1 to 6.

10. A computer program product, comprising computer program instructions, which, when executed on a computer, cause the computer to execute the method according to any one of claims 1 to 6.

Images