A power customer service-based operation analysis method and system

Abstract

The application relates to the technical field of operation analysis, and particularly discloses an operation analysis method and system based on power customer service, which comprises the following steps: reading customer service data containing customer historical service records obtained by a customer service terminal according to time sequence, and classifying the customer service data according to the customer historical service records; analyzing the customer service data containing sender marks to obtain data such as traffic volume, service duration, manual transfer rate and recognition rate in the customer service data; and generating an operation analysis report by randomly selecting customer service data based on the data such as traffic volume, service duration, manual transfer rate and recognition rate. The customer service data is converted into text, and the text is analyzed by means of text recognition technology; then, each conversation is randomly selected according to a preset quantity, and keywords in each conversation are improved according to a preset keyword library, so that an operation analysis report is generated from the keywords and the repetition times of the keywords, and the customer service data is effectively utilized.

Description

Technical Field

[0001] This invention relates to the field of operational analysis technology, specifically an operational analysis method and system based on power customer service. Background Technology

[0002] Intelligent customer service is an industry-oriented technology developed on the basis of large-scale knowledge processing. It is based on (large-scale knowledge processing technology, natural language understanding technology, knowledge management technology, automatic question answering system, reasoning technology, etc.) and has industry universality. It not only provides enterprises with fine-grained knowledge management technology, but also establishes a fast and effective technical means for communication between enterprises and a large number of users based on natural language. At the same time, it can also provide enterprises with the statistical analysis information needed for refined management.

[0003] In the field of power services, intelligent customer service technology is also applied. When users encounter power problems, they can communicate directly with intelligent customer service. During the communication process, users are likely to solve some minor problems. Once the problem is solved, the customer will no longer communicate with the AI. The system will store and back up the communication process and then delete it periodically. In fact, this data is very valuable for analysis, and existing technologies almost completely ignore this data. Summary of the Invention

[0004] The purpose of this invention is to provide an operational analysis method and system based on power customer service to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] An operational analysis method based on electricity customer service, the method comprising:

[0007] Customer service data containing customer history service records obtained from the client is read in chronological order, and the customer service data is classified according to the customer history service records; the customer service data contains sender tags;

[0008] Analyze customer service data containing sender tags to determine an emotion score; wherein the dependent variables of the emotion score include call volume, service duration, transfer rate to human agent, and recognition rate in the customer service data;

[0009] Customer service data is randomly selected based on sentiment scores. Keywords and their frequencies are extracted from the customer service data to generate a keyword table.

[0010] The keyword list is identified to generate an operational analysis report.

[0011] As a further aspect of the present invention: the step of reading customer service data containing customer historical service records obtained from the customer service terminal in chronological order, and classifying the customer service data according to the customer historical service records includes:

[0012] The switch signals of all clients are acquired in real time; the switch signals are edge-triggered signals.

[0013] Based on the same time axis, the switching signals are statistically analyzed, the total value of the switching signals at each moment is calculated sequentially, and the time period is determined based on the total value of the switching signals.

[0014] The system uses time periods as labels to statistically analyze customer service data containing historical service records, and then categorizes the customer service data based on these historical service records.

[0015] As a further aspect of the present invention: the step of analyzing customer service data containing sender tags to determine the sentiment score includes:

[0016] The sender text is obtained by classifying the terms in the customer service data according to the sender tags in the customer service data.

[0017] The sender's text is extracted sequentially according to a preset incremental step size to obtain the text to be inspected containing a length label; the length label is the proportion of the text to be inspected to the sender's text.

[0018] The text to be examined is subjected to part-of-speech tagging and facial expression recognition, and emotional words and emotional expressions are marked.

[0019] The customer's emotion score is calculated in real time based on the marked emotion words and emotions, and the emotion score is sorted according to the length label of the text to be inspected.

[0020] The sorted sentiment scores are analyzed, and the feedback data is labeled; the feedback data is generated by the intelligent customer service terminal.

[0021] As a further aspect of the present invention: the step of analyzing the sorted emotion scores and labeling the feedback data includes:

[0022] Query the length label of the text to be inspected corresponding to each emotion score, and convert the emotion scores into a score curve based on the length label;

[0023] Calculate the derivative of the scoring curve, compare the derivative with a preset derivative threshold, and mark the corresponding length label when the derivative reaches the preset derivative threshold;

[0024] Centered on the length label, query the target text in the customer service data, and provide feedback data based on the location of the target text.

[0025] As a further aspect of the present invention: the step of randomly selecting customer service data based on emotion scores, extracting keywords and their frequencies from the customer service data, and generating a keyword table includes:

[0026] Customer service data is categorized based on emotion scores, and customer service data is selected from each category according to a preset selection quantity. Different emotion scores correspond to an emotion level, and each emotion level corresponds to a selection quantity.

[0027] Based on a pre-defined keyword database, keywords are extracted from customer service data and their frequencies are calculated.

[0028] Analyze keywords and their frequencies to generate a keyword table.

[0029] As a further aspect of the present invention: the step of identifying the keyword list and generating an operational analysis report includes:

[0030] Normalize each keyword in the keyword table; the normalization process involves querying synonyms for each keyword in a preset thesaurus, arranging the synonyms according to a preset order, and selecting the first synonym as the normalization result.

[0031] Merge the keyword lists after normalization to obtain the unified keyword list;

[0032] Input the vocabulary list into the trained report generation model to generate an operational analysis report;

[0033] The report generation model is a word-analysis report mapping model.

[0034] The present invention also provides an operation analysis system based on power customer service, the system comprising:

[0035] The customer service data classification module is used to read customer service data containing customer history service records obtained from the customer service terminal in chronological order, and classify the customer service data according to the customer history service records; the customer service data contains sender tags;

[0036] The emotion scoring module is used to analyze customer service data containing sender tags to determine the emotion score; wherein, the dependent variables of the emotion score include call volume, service duration, transfer rate to human agent and recognition rate in the customer service data;

[0037] The keyword extraction module is used to randomly select customer service data based on sentiment scores, extract keywords and their frequencies from the customer service data, and generate a keyword table.

[0038] The report generation module is used to identify the keyword list and generate an operational analysis report.

[0039] As a further aspect of the present invention: the customer service data classification module includes:

[0040] A switch signal acquisition unit is used to acquire the switch signals of all clients in real time; the switch signal is a transition edge signal.

[0041] The time period determination unit is used to statistically analyze the switching signals based on the same time axis, calculate the total value of the switching signals at each moment in sequence, and determine the time period based on the total value of the switching signals.

[0042] The statistical classification unit is used to statistically analyze customer service data containing customer history service records by time period label, and to classify the customer service data according to the customer history service records.

[0043] As a further aspect of the present invention: the emotion scoring module includes:

[0044] The term classification unit is used to classify each term in the customer service data according to the sender's marker in the customer service data, and obtain the sender's text;

[0045] The text truncation unit is used to truncate the sender's text sequentially according to a preset incremental step size to obtain the text to be inspected containing a length label; the length label is the proportion of the text to be inspected to the sender's text.

[0046] The text recognition unit is used to perform part-of-speech recognition and expression recognition on the text to be inspected, and to mark emotion words and emotion expressions.

[0047] The score sorting unit is used to calculate the customer's emotion score in real time based on the marked emotion words and emotion expressions, and sort the emotion scores according to the length label of the text to be inspected.

[0048] The feedback data labeling unit is used to analyze the sorted emotion scores and label the feedback data; the entity that generates the feedback data is the intelligent customer service terminal.

[0049] As a further aspect of the present invention: the keyword extraction module includes:

[0050] The data selection unit is used to classify customer service data according to emotion scores and select customer service data from each category according to a preset selection quantity. Different emotion scores correspond to an emotion level, and an emotion level corresponds to a selection quantity.

[0051] The traversal extraction unit is used to traverse and extract keywords from customer service data based on a preset keyword library and calculate their frequency.

[0052] The statistical generation unit is used to count keywords and their frequencies, and generate a keyword table.

[0053] Compared with the prior art, the beneficial effects of the present invention are as follows: The present invention converts customer service data into text, analyzes the text with the help of text recognition technology, determines the user's emotions in real time, and then clusters each conversation; then, randomly selects each conversation according to a preset number, improves the keywords in each conversation according to a preset keyword library, and generates an operation analysis report from the keywords and their repetition frequency, effectively utilizing customer service data. Attached Figure Description

[0054] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention.

[0055] Figure 1 This is a flowchart of an operational analysis method based on electricity customer service.

[0056] Figure 2 This is the first sub-process flowchart of the operational analysis method based on electricity customer service.

[0057] Figure 3 This is the second sub-process flowchart of the operational analysis method based on electricity customer service.

[0058] Figure 4 This is the third sub-process flowchart of the operational analysis method based on electricity customer service.

[0059] Figure 5 This is the fourth sub-process flowchart of the operational analysis method based on electricity customer service.

[0060] Figure 6 This is a block diagram of the structural composition of an operation analysis system based on electricity customer service. Detailed Implementation

[0061] To make the technical problems to be solved, the technical solutions, and the beneficial effects of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the present invention and are not intended to limit the present invention.

[0062] Example 1

[0063] Figure 1 The flowchart below illustrates an operational analysis method based on electricity customer service. In this embodiment of the invention, an operational analysis method based on electricity customer service includes:

[0064] Step S100: Read the customer service data containing customer history service records obtained from the customer service terminal in chronological order, and classify the customer service data according to the customer history service records; the customer service data contains sender tags;

[0065] The customer service data in the technical solution of this invention is the dialogue between the visitor and the intelligent customer. During the dialogue, the subject of this method will obtain the customer's historical service records. In the field of power application, the information of each user is saved. Therefore, it is not difficult to obtain the customer's historical service records.

[0066] Step S200: Analyze the customer service data containing sender tags to determine the emotion score; wherein, the dependent variables of the emotion score include call volume, service duration, transfer to human agent rate, and recognition rate in the customer service data;

[0067] Sender tags are used to indicate which party sent a particular term; analyzing customer service data containing sender tags can determine sentiment scores.

[0068] Step S300: Randomly select customer service data based on sentiment scores, extract keywords and their frequencies from the customer service data, and generate a keyword table;

[0069] Emotion scores are used to characterize the stability of a user's emotions, and the identification value of customer service data varies depending on the emotion score. Therefore, by selecting different amounts of customer service data based on emotion scores and then analyzing the selected customer service data, a keyword table can be obtained.

[0070] Step S400: Identify the keyword list and generate an operational analysis report;

[0071] By identifying the keywords, an operational analysis report can be obtained.

[0072] Figure 2 The first sub-process flowchart of the operation analysis method based on power customer service includes the step of reading customer service data containing customer historical service records obtained from the customer service terminal in chronological order, and classifying the customer service data according to the customer historical service records.

[0073] Step S101: Acquire the switching signals of all clients in real time; the switching signals are edge-triggered signals;

[0074] Whether each customer service representative is in operation can be monitored in real time by the executing entity of the method, and the resulting switch status is the aforementioned switch signal; it can be imagined that the switch signal is a transition edge signal.

[0075] Step S102: Based on the same time axis, statistically analyze the switching signals, calculate the total value of the switching signals at each moment in sequence, and determine the time period based on the total value of the switching signals;

[0076] Based on the same timeline, all client switch signals are statistically analyzed, and the total value of the switch signals at each time point is calculated (the total value has no physical meaning). The higher the total value, the more clients are running at that time, and correspondingly, the time period used for classification should be shorter.

[0077] Step S103: Collect customer service data containing customer history service records by time period label, and classify the customer service data according to the customer history service records;

[0078] Customer service data is statistically analyzed using time periods as tags, and then categorized based on customers' historical service records. The logic of this process is to first categorize the customer service data based on time, and then perform a second categorization based on location.

[0079] Figure 3 The second sub-process flowchart of the operational analysis method based on electricity customer service includes the step of analyzing customer service data containing sender tags to determine sentiment scores, which includes:

[0080] Step S201: Classify each term in the customer service data according to the sender tag in the customer service data to obtain the sender text;

[0081] The customer service data is categorized based on the sender's tags to obtain the sender's text, which is a set of terms sent by the user.

[0082] Step S202: Extract the sender's text sequentially according to a preset incremental step size to obtain the text to be inspected containing a length tag; the length tag is the proportion of the text to be inspected to the sender's text.

[0083] The sender's text is truncated to obtain text to be inspected of different lengths, and the length is represented by a length label.

[0084] Step S203: Perform part-of-speech and expression recognition on the text to be inspected, and mark emotion words and emotion expressions;

[0085] The text to be inspected contains words and expressions. The process of part-of-speech recognition is not difficult and can draw on existing text recognition technologies. The result of part-of-speech recognition and expression recognition is to obtain emotional words and emotional expressions. The emotional words are preset by the user. For example, "haha" is considered as happy and the emotion is relatively happy; "OK" is considered as calm and the emotion is relatively calm. The specific determination is made by the staff according to the situation.

[0086] Step S204: Calculate the customer's emotion score in real time based on the marked emotion words and emotion expressions, and sort the emotion scores according to the length label of the text to be inspected;

[0087] Each emotion word and emotion expression reflects the customer's emotions. According to the preset conversion relationship, the emotion words and emotion expressions are converted into a score. Each score corresponds to a text to be inspected. The emotion scores are sorted according to the length of the text to be inspected.

[0088] Step S205: Analyze the sorted sentiment scores and label the feedback data; the feedback data is generated by the intelligent customer service terminal.

[0089] The sorted emotion scores are analyzed, and the time periods of significant emotion changes are located based on the analysis results. Then, the feedback data generated by the subject executing this method is retrieved.

[0090] As a preferred embodiment of the technical solution of the present invention, the step of analyzing the sorted emotion scores and labeling the feedback data includes:

[0091] Query the length label of the text to be inspected corresponding to each emotion score, and convert the emotion scores into a score curve based on the length label;

[0092] Calculate the derivative of the scoring curve, compare the derivative with a preset derivative threshold, and mark the corresponding length label when the derivative reaches the preset derivative threshold;

[0093] The rating curve represents the changes in users' emotions, and the derivative of the rating curve represents the rate of change in emotions. When the rate of change reaches a certain level, it indicates that the user's emotions have changed drastically for that particular term. The reason for this drastic change is likely due to feedback data, so it is necessary to query the feedback data.

[0094] Centered on the length label, query the target text in the customer service data, and provide feedback data based on the position of the target text;

[0095] The target text is queried by the length tag. The target text is the content sent by the client. The content sent by the query method execution end is centered on the position of the target text, which is the feedback data.

[0096] Figure 4 The third sub-process flowchart of the operation analysis method based on electricity customer service includes the following steps: randomly selecting customer service data based on sentiment scores, extracting keywords and their frequencies from the customer service data, and generating a keyword table:

[0097] Step S301: Classify customer service data according to emotion scores, and select customer service data from each category according to a preset selection quantity; wherein, different emotion scores correspond to an emotion level, and an emotion level corresponds to a selection quantity.

[0098] Emotion scores are categorized to determine the amount of customer service data for each score range, and customer service data is randomly selected from all customer service data.

[0099] Step S302: Extract keywords from customer service data based on a preset keyword database and calculate their frequency;

[0100] Step S303: Count keywords and their frequencies, and generate a keyword table.

[0101] By using existing keyword extraction technology to identify customer service data, a keyword table can be obtained.

[0102] Figure 5 The fourth sub-process flowchart of the operation analysis method based on electricity customer service includes the following steps: identifying the keyword table and generating an operation analysis report.

[0103] Step S401: Normalize each keyword in the keyword table; the normalization process involves querying synonyms for each keyword in a preset thesaurus, arranging the synonyms according to a preset order, and selecting the first synonym as the normalization result.

[0104] Step S402: Merge the normalized keyword lists to obtain a unified keyword list;

[0105] Step S403: Input the vocabulary list into the trained report generation model to generate an operational analysis report;

[0106] The report generation model is a word-analysis report mapping model.

[0107] The keyword table contains keywords and their frequency entries. Many keywords are actually synonyms. Therefore, by performing synonym conversion on the keywords to obtain identical words, and merging the frequencies of identical words, a unified keyword table can be obtained. In the preset database, the analysis report is queried based on the words in the unified keyword table and their occurrence frequency.

[0108] Example 2

[0109] Figure 6 The diagram shows the structural composition of an operation analysis system based on electricity customer service. In this embodiment of the invention, an operation analysis system based on electricity customer service, system 10, includes:

[0110] The customer service data classification module 11 is used to read customer service data containing customer history service records obtained from the customer service terminal in chronological order, and classify the customer service data according to the customer history service records; the customer service data contains sender tags;

[0111] The emotion scoring module 12 is used to analyze customer service data containing sender tags and determine the emotion score; wherein, the dependent variables of the emotion score include call volume, service duration, transfer to human agent rate and recognition rate in the customer service data;

[0112] Keyword extraction module 13 is used to randomly select customer service data based on sentiment scores, extract keywords and their frequencies from the customer service data, and generate a keyword table;

[0113] The report generation module 14 is used to identify the keyword list and generate an operational analysis report.

[0114] The customer service data classification module 11 includes:

[0115] A switch signal acquisition unit is used to acquire the switch signals of all clients in real time; the switch signal is a transition edge signal.

[0116] The time period determination unit is used to statistically analyze the switching signals based on the same time axis, calculate the total value of the switching signals at each moment in sequence, and determine the time period based on the total value of the switching signals.

[0117] The statistical classification unit is used to statistically analyze customer service data containing customer history service records by time period label, and to classify the customer service data according to the customer history service records.

[0118] The emotion scoring module 12 includes:

[0119] The term classification unit is used to classify each term in the customer service data according to the sender's marker in the customer service data, and obtain the sender's text;

[0120] The text truncation unit is used to truncate the sender's text sequentially according to a preset incremental step size to obtain the text to be inspected containing a length label; the length label is the proportion of the text to be inspected to the sender's text.

[0121] The text recognition unit is used to perform part-of-speech recognition and expression recognition on the text to be inspected, and to mark emotion words and emotion expressions.

[0122] The score sorting unit is used to calculate the customer's emotion score in real time based on the marked emotion words and emotion expressions, and sort the emotion scores according to the length label of the text to be inspected.

[0123] The feedback data labeling unit is used to analyze the sorted emotion scores and label the feedback data; the entity that generates the feedback data is the intelligent customer service terminal.

[0124] The keyword extraction module 13 includes:

[0125] The data selection unit is used to classify customer service data according to emotion scores and select customer service data from each category according to a preset selection quantity. Different emotion scores correspond to an emotion level, and an emotion level corresponds to a selection quantity.

[0126] The traversal extraction unit is used to traverse and extract keywords from customer service data based on a preset keyword library and calculate their frequency.

[0127] The statistical generation unit is used to count keywords and their frequencies, and generate a keyword table.

[0128] The functions of the operation analysis method based on electricity customer service are all performed by computer equipment, which includes one or more processors and one or more memories. The one or more memories store at least one piece of program code, which is loaded and executed by the one or more processors to realize the functions of the operation analysis method based on electricity customer service.

[0129] The processor fetches instructions from memory one by one, analyzes the instructions, and then performs the corresponding operations according to the instructions, generating a series of control commands to enable the various parts of the computer to act automatically, continuously, and in a coordinated manner, forming an organic whole. This enables the input of programs and data, as well as the calculation and output of results. The arithmetic or logical operations generated in this process are all performed by the arithmetic unit. The memory includes a read-only memory (ROM), which is used to store computer programs. The memory is protected by an external protection device.

[0130] For example, a computer program can be divided into one or more modules, one or more of which are stored in memory and executed by a processor to perform the present invention. The one or more modules can be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in a terminal device.

[0131] Those skilled in the art will understand that the above description of the service equipment is merely an example and does not constitute a limitation on the terminal equipment. It may include more or fewer components than described above, or combine certain components, or different components, such as input / output devices, network access devices, buses, etc.

[0132] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. This processor is the control center of the terminal device, connecting various parts of the user terminal via various interfaces and lines.

[0133] The aforementioned memory can be used to store computer programs and / or modules. The aforementioned processor implements various functions of the aforementioned terminal device by running or executing the computer programs and / or modules stored in the memory, and by calling data stored in the memory. The memory may mainly include a program storage area and a data storage area. The program storage area may store the operating system, at least one application program required for a function (such as information collection template display function, product information publishing function, etc.); the data storage area may store data created based on the use of the berth status display system (such as product information collection templates corresponding to different product types, product information that different product providers need to publish, etc.). In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.

[0134] If the modules / units integrated into the terminal device are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, all or part of the modules / units in the systems of the above embodiments can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the functions of the various system embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc.

[0135] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

[0136] The above are merely preferred embodiments of the present invention and do not limit the scope of the patent. Any equivalent structural or procedural transformations made based on the description and drawings of the present invention, or direct or indirect applications in other related technical fields, are similarly included within the scope of patent protection of the present invention.

Claims

1. An operational analysis method based on electricity customer service, characterized in that, The method includes: Customer service data containing visitor locations obtained from the client is read in chronological order, and the customer service data is classified according to the visitor locations; the customer service data contains sender tags; Analyze customer service data containing sender tags to determine sentiment scores; Customer service data is randomly selected based on sentiment scores. Keywords and their frequencies are extracted from the customer service data to generate a keyword table. The keyword list is identified to generate an operational analysis report; Customer service data is categorized based on emotion scores, and customer service data is selected from each category according to a preset selection quantity. Different emotion scores correspond to different emotion levels, and each emotion level corresponds to a selection quantity. Emotion scores are graded to determine the amount of customer service data for each score range, and customer service data is randomly selected from all customer service data. Keywords are extracted from the customer service data based on a preset keyword library, and their frequencies are calculated. Keywords and their frequencies are statistically analyzed to generate a keyword table. The keywords in the keyword table are normalized. The normalization process involves querying synonyms for each keyword in a preset thesaurus, arranging the synonyms according to a preset order, and selecting the first synonym as the normalization result. The normalized keyword tables are then merged to obtain a unified word table. The unified word table is then input into a trained report generation model to generate an operational analysis report.

2. The operational analysis method based on electricity customer service according to claim 1, characterized in that, The step of reading customer service data containing visitor locations from the customer service terminal in chronological order and classifying the customer service data according to the visitor locations includes: The switch signals of all clients are acquired in real time; the switch signals are edge-triggered signals. Based on the same time axis, the switching signals are statistically analyzed, the total value of the switching signals at each moment is calculated sequentially, and the time period is determined based on the total value of the switching signals. The system uses time periods as tags to collect customer service data containing visitor locations, and then categorizes the customer service data based on visitor locations.

3. The operational analysis method based on electricity customer service according to claim 1, characterized in that, The step of analyzing customer service data containing sender tags to determine the sentiment score includes: The sender text is obtained by classifying the terms in the customer service data according to the sender tags in the customer service data. The sender's text is extracted sequentially according to a preset incremental step size to obtain the text to be inspected containing a length label; the length label is the proportion of the text to be inspected to the sender's text. The text to be examined is subjected to part-of-speech tagging and facial expression recognition, and emotional words and emotional expressions are marked. The customer's emotion score is calculated in real time based on the marked emotion words and emotions, and the emotion score is sorted according to the length label of the text to be inspected. The sorted sentiment scores are analyzed, and the feedback data is labeled; the feedback data is generated by the intelligent customer service terminal.

4. The operational analysis method based on electricity customer service according to claim 3, characterized in that, The steps of analyzing the sorted emotion scores and labeling the feedback data include: Query the length label of the text to be inspected corresponding to each emotion score, and convert the emotion scores into a score curve based on the length label; Calculate the derivative of the scoring curve, compare the derivative with a preset derivative threshold, and mark the corresponding length label when the derivative reaches the preset derivative threshold; Centered on the length label, query the target text in the customer service data, and provide feedback data based on the location of the target text.

5. An operational analysis system based on electricity customer service, characterized in that, The system includes: The customer service data classification module is used to read customer service data containing visitor locations obtained from the customer service terminal in chronological order, and classify the customer service data according to the visitor locations; the customer service data contains sender tags; The emotion scoring module is used to analyze customer service data containing sender tags and determine the emotion score. The keyword extraction module is used to randomly select customer service data based on sentiment scores, extract keywords and their frequencies from the customer service data, and generate a keyword table. The report generation module is used to identify the keyword list and generate an operational analysis report; Customer service data is categorized based on emotion scores, and customer service data is selected from each category according to a preset selection quantity. Different emotion scores correspond to different emotion levels, and each emotion level corresponds to a selection quantity. Emotion scores are graded to determine the amount of customer service data for each score range, and customer service data is randomly selected from all customer service data. Keywords are extracted from the customer service data based on a preset keyword library, and their frequencies are calculated. Keywords and their frequencies are statistically analyzed to generate a keyword table. The keywords in the keyword table are normalized. The normalization process involves querying synonyms for each keyword in a preset thesaurus, arranging the synonyms according to a preset order, and selecting the first synonym as the normalization result. The normalized keyword tables are then merged to obtain a unified word table. The unified word table is then input into a trained report generation model to generate an operational analysis report.

6. The operation analysis system based on electricity customer service according to claim 5, characterized in that, The customer service data classification module includes: A switch signal acquisition unit is used to acquire the switch signals of all clients in real time; the switch signal is a transition edge signal. The time period determination unit is used to statistically analyze the switching signals based on the same time axis, calculate the total value of the switching signals at each moment in sequence, and determine the time period based on the total value of the switching signals. The statistical classification unit is used to statistically analyze customer service data containing visitor locations by time period label, and to classify the customer service data according to visitor location.

7. The operation analysis system based on electricity customer service according to claim 5, characterized in that, The emotion scoring module includes: The term classification unit is used to classify each term in the customer service data according to the sender's marker in the customer service data, and obtain the sender's text; The text truncation unit is used to truncate the sender's text sequentially according to a preset incremental step size to obtain the text to be inspected containing a length label; the length label is the proportion of the text to be inspected to the sender's text. The text recognition unit is used to perform part-of-speech recognition and expression recognition on the text to be inspected, and to mark emotion words and emotion expressions. The score sorting unit is used to calculate the customer's emotion score in real time based on the marked emotion words and emotion expressions, and sort the emotion scores according to the length label of the text to be inspected. The feedback data labeling unit is used to analyze the sorted emotion scores and label the feedback data; the entity that generates the feedback data is the intelligent customer service terminal.

Images