Enterprise management consultation customer service allocation method based on question characteristics

Abstract

The invention relates to the technical field of enterprise management consultation. The invention relates to an enterprise management consultation customer service allocation method based on question characteristics. The method comprises the following steps: S1, collecting historical problem solving records of customer service staff, constructing an enterprise management knowledge graph in combination with the historical problem solving records, and setting the responsible field of each customer service staff; s2, performing problem solving quality analysis according to historical problem solving records, obtaining the solving quality of the problem solving records, and performing problem feature and corresponding reply utterance retention on the problem solving records with high solving quality; according to the method, client problem data is collected in real time, problem field analysis is carried out in combination with an enterprise management knowledge graph, the field to which the problem belongs can be rapidly and accurately determined, for example, whether the problem is a supply chain problem, a marketing promotion problem or a product after-sales problem is identified, and then a similar threshold value is dynamically set according to the problem feature complexity difficulty; and accurately matching with historical problem solving records.

Description

Technical Field

[0001] The present invention relates to the technical field of enterprise management consulting, and in particular to an enterprise management consulting customer service allocation method based on problem characteristics. Background Art

[0002] The customer service allocation model of enterprise management consulting mostly operates under a relatively fixed process framework. When customers report problems through channels such as phone, email or online platforms, customer service staff usually take on tasks in a predetermined order.

[0003] However, enterprise problems often involve the intersection and integration of multiple business areas. For example, when enterprises implement digital transformation, it involves many issues such as information technology, business process reshaping, and personnel training. Traditional classification methods can easily lead to inaccurate problem allocation, causing customer service personnel to face problems beyond their professional capabilities. This not only prolongs customer waiting time and reduces customer satisfaction, but may also affect the corporate image due to unprofessional answers. Therefore, a method for allocating customer service for enterprise management consulting based on problem characteristics is proposed. Summary of the invention

[0004] The purpose of the present invention is to provide a method for allocating enterprise management consulting customer service based on problem characteristics to solve the problems raised in the above background technology.

[0005] To achieve the above purpose, a method for allocating customer service for enterprise management consulting based on problem characteristics is provided, comprising the following steps:

[0006] S1. Collect the historical problem solving records of customer service, and build the enterprise management knowledge map based on the historical problem solving records, and set the responsible areas of each customer service;

[0007] S2. Analyze the problem solving quality based on historical problem solving records to obtain the problem solving quality of the problem solving records, and retain the problem features and corresponding response discourse of the problem solving records with high solution quality;

[0008] S3. Combine the problem solving records with the problem characteristics and solution quality to generate a customer service capability profile for the assigned customer service. Then collect the customer's problem data and combine the problem data with the enterprise management knowledge graph to perform problem domain analysis. At the same time, perform problem characteristic analysis on the problem data.

[0009] S4. Perform similar matching on the problem features of the problem data and the problem features of the problem solving record, and display the corresponding reply utterance to the customer according to the matching result, and assign the customer to the corresponding customer service of the reply utterance;

[0010] S5. When S4 does not meet the matching requirements, the problem characteristics of the problem data are combined with the customer service capability profile and satisfaction indicators that match the problem field to conduct the best customer service analysis, and the customer is assigned to the corresponding customer service according to the analysis results.

[0011] As a further improvement of the present technical solution, S1 is connected to a port managed by the enterprise, and then collects historical data related to customers from the port, and extracts historical problem solving records from the historical data.

[0012] As a further improvement of the technical solution, the step of S1 is as follows:

[0013] S1.1. Collect the business areas of the enterprise, and then combine the business areas of the enterprise with the historical problem solving records to construct the enterprise management knowledge graph in a structured manner, and use the enterprise management knowledge graph to determine the problem areas to which the historical problem solving records belong;

[0014] S1.2. Collect the list of customer service staff and historical problem solving records to determine the problem areas that each customer service representative has solved, and set them as the customer service representative’s area of ​​responsibility.

[0015] As a further improvement of the technical solution, the steps of S2 are as follows:

[0016] S2.1. Split the overall historical problem solving records according to the problems and obtain multiple different historical problem solving records;

[0017] S2.2. Set up a satisfaction rating mechanism. The satisfaction rating is divided into high, medium, and low. At the same time, extract the problem features and corresponding reply discourses from multiple different historical problem solving records. Then, combine the problem features and corresponding reply discourses with the customer satisfaction ratings for quality analysis. When the problem features match the reply discourses and the customer satisfaction rating is high, the problem solving record is judged to be of high resolution quality, and the problem features and corresponding reply discourses are saved. Otherwise, they are not saved.

[0018] As a further improvement of the technical solution, the steps of S3 are as follows:

[0019] S3.1. Generate a customer service capability profile based on the problem solving records, problem characteristics, and solution quality of each customer service representative, and obtain a customer service capability profile for each customer service representative;

[0020] S3.2. Collect customer problem data in real time from the enterprise management port, and combine the problem data with the enterprise management knowledge graph to perform problem domain analysis to obtain the problem domain corresponding to the problem data. At the same time, perform problem feature analysis on the problem data to obtain the problem features corresponding to the problem data.

[0021] As a further improvement of the technical solution, the step of S4 is as follows:

[0022] S4.1. Perform complexity and difficulty analysis on the problem features of the problem data, and dynamically set similarity thresholds based on the complexity and difficulty;

[0023] When the problem features are more complex and difficult, the similarity threshold is lower;

[0024] When the problem features are more complex and less difficult, the similarity threshold is higher;

[0025] S4.2. The problem features of the problem data are combined with the problem features of the problem solution record for similarity matching. When the similarity between the problem features of the problem data and the problem features of the problem solution record meets the similarity threshold, the reply utterance corresponding to the problem features is displayed to the customer, and the customer is assigned to the corresponding customer service of the reply utterance. Conversely, when the similarity between the problem features of the problem data and the problem features of the problem solution record does not meet the similarity threshold, enter S5.

[0026] As a further improvement of the technical solution, the S4.2 further includes the following steps:

[0027] S4.2.1. When there are multiple problem features in the problem data, the problem features are compared in the problem domain. When all the problem features are in the same problem domain, the customer service with the highest similarity among all the problem feature matches is selected for assignment;

[0028] S4.2.2. When there are multiple problem features in the problem data, and the multiple problem features do not belong to the same problem field, the problem features are sorted according to the problem data, and the customer service with the highest similarity of the first problem feature is assigned.

[0029] As a further improvement of the technical solution, the step of S5 is as follows:

[0030] S5.1. Use recent response time, recent customer satisfaction score, and the number of people being received as satisfaction indicators;

[0031] S5.2: When the similarity of all problem features in S4.2 does not meet the similarity threshold, the problem features of the problem data are combined with the customer service capability profile and satisfaction index that meet the problem domain to conduct the best customer service analysis, and the customer is assigned to the corresponding customer service according to the analysis results.

[0032] Compared with the prior art, the present invention has the following beneficial effects:

[0033] 1. In this enterprise management consulting customer service allocation method based on problem characteristics, by collecting customer problem data in real time and combining it with the enterprise management knowledge graph to analyze the problem field, it can quickly and accurately determine the field to which the problem belongs, such as identifying whether it is a supply chain problem, a marketing promotion problem, or a product after-sales problem, and then dynamically set the similarity threshold based on the complexity and difficulty of the problem characteristics, and accurately match it with the historical problem solving records to ensure that customer problems can be directed to the customer service personnel who are most familiar with the corresponding field and most likely to provide effective solutions. This greatly shortens the problem solving cycle, improves the effectiveness of the first response, reduces customer waiting time, and improves customer satisfaction.

[0034] 2. In this method of allocating customer service for enterprise management consulting based on problem characteristics, by generating a customer service capability portrait based on the problem solving records, problem characteristics and solution quality handled by the customer service, the management can clearly understand the performance of each customer service representative in different fields, such as their professional knowledge mastery, problem solving efficiency, communication skills and cross-domain collaboration capabilities. On the one hand, targeted training plans can be formulated for customer service personnel to improve their weak skills; on the other hand, when allocating tasks, the customer service personnel's strengths can be fully utilized, resources can be reasonably allocated, and the work efficiency of the entire customer service team can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 It is the overall flow chart of the present invention;

[0036] Figure 2 A flowchart of collecting a list of customer service personnel for the present invention;

[0037] Figure 3 A flowchart for obtaining multiple different historical problem solving records for the present invention;

[0038] Figure 4 A flowchart of obtaining a customer service capability portrait of each customer service representative in the present invention;

[0039] Figure 5 A flowchart of dynamically setting a similarity threshold according to complexity and difficulty of the present invention;

[0040] Figure 6 This is a flowchart of the present invention for allocating customers to corresponding customer service according to the analysis results. DETAILED DESCRIPTION

[0041] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0042] See also Figure 1 - Figure 6 As shown, the purpose of this embodiment is to provide a method for allocating customer service for enterprise management consulting based on problem characteristics, including the following steps:

[0043] S1. Collect the historical problem solving records of customer service, and build the enterprise management knowledge map based on the historical problem solving records, and set the responsible areas of each customer service;

[0044] S1 connects to the port managed by the enterprise, collects customer-related historical data from the port, and extracts historical problem-solving records from the historical data.

[0045] The steps of S1 are as follows:

[0046] S1.1. Collect the business areas of the enterprise, and then combine the business areas of the enterprise with the historical problem solving records to construct the enterprise management knowledge graph in a structured manner. Use the enterprise management knowledge graph to determine the problem areas to which the historical problem solving records belong. The specific steps are as follows:

[0047] Collecting the business areas of the enterprise: Analyze the enterprise's past project documents, work reports, etc., find out the frequently appearing keywords of business activities, summarize them into business areas, or have the heads of various departments within the enterprise compile a list of business areas;

[0048] Build an enterprise management knowledge graph: Identify entities such as business departments, business processes, and management issues from keywords in the enterprise business field and historical problem-solving records, and then extract the relationships between entities, such as a business department is responsible for a business process, a management problem affects a business field, etc. Then use a graph database (such as Neo4j) or a knowledge graph building tool to store the identified entities and relationships in a structured manner to form an enterprise management knowledge graph, with each entity as a node and the relationships between entities as edges;

[0049] Determine the problem domain to which historical problem-solving records belong: Perform text analysis on the problem descriptions in historical problem-solving records, extract key information and keywords, match the extracted keywords with entities in the enterprise management knowledge graph, and then manually review the automatically determined problem domain to check for errors or inaccurate classifications. If problems are found, you can manually adjust the domain to which the problem-solving records belong, and optimize the knowledge graph and matching algorithm.

[0050] S1.2. Collect the list of customer service personnel and historical problem solving records to determine the problem areas that each customer service has solved, and set them as the customer service's responsible areas. The specific steps are as follows:

[0051] Collect customer service personnel list: By accessing the module in the human resources management system (HRMS) that specifically records customer service personnel information, extract the basic information of all active customer service personnel, including name, work number, etc., to form a complete list of customer service personnel

[0052] If the enterprise has a dedicated customer service work platform (such as an online customer service system, a customer service ticket system, etc.), it can search and summarize the information of all personnel with customer service authority and actually involved in handling customer issues from the user management module of the platform, and add them to the customer service personnel list to ensure that the list covers all customer service individuals involved in handling enterprise management consulting issues;

[0053] Organize historical problem solving records: Extract all past enterprise management consulting problem solving records from the relevant systems or databases that store historical data (such as customer service ticket databases, problem record archiving systems, etc.). These records should include key information such as a detailed description of the problem, processing time, identification of the customer service staff responsible for handling the problem (such as name or work number), and the final resolution of the problem;

[0054] Determine the problem domains solved by each customer service representative: For the problem description in each historical problem solving record, use natural language processing technology (such as lexical analysis tools, TF-IDF algorithm, etc.) to extract keywords that can represent the domain to which the problem belongs, match the extracted keywords with this classification system, and then determine the domain to which each problem belongs. According to the responsible customer service representative ID indicated in each historical problem solving record, associate the determined problem domain with the corresponding customer service representative;

[0055] Set the customer service area of ​​responsibility: For each customer service staff, count the frequency of their handling of problems in various problem areas, set a frequency threshold, and for each customer service staff, set the problem area with a handling frequency reaching the threshold and with a relatively high frequency as their area of ​​responsibility.

[0056] S2. Analyze the problem solving quality based on historical problem solving records to obtain the problem solving quality of the problem solving records, and retain the problem features and corresponding response discourse of the problem solving records with high solution quality;

[0057] The steps of S2 are as follows:

[0058] S2.1. Split the overall historical problem solving records according to the problems and obtain multiple different historical problem solving records;

[0059] Take the problem as the unit and split the different problems involved in each record so that each split record corresponds to only one specific problem. For example, if an original record contains two different problems that the customer asked successively and the corresponding answers, it will be split into two independent historical problem solving records, each record focusing on a separate problem. The splitting operation can be performed by identifying the separation mark in the problem description (such as other conjunctions, or by analyzing the conversion of the problem topic, etc.).

[0060] S2.2. Set up a satisfaction rating mechanism. The satisfaction rating is divided into high, medium and low. At the same time, extract the problem features and corresponding reply words from multiple different historical problem solving records. Then, combine the problem features and corresponding reply words with the customer satisfaction score for quality analysis. When the problem features match the reply words and the customer satisfaction score is high, the problem solving record is judged to have high solution quality. The problem features and corresponding reply words are saved. Otherwise, they are not saved. The specific steps are as follows:

[0061] Determine the scoring criteria: Develop specific satisfaction scoring criteria based on the actual situation and past experience of the enterprise. For example, the criteria for setting a high satisfaction score is that the customer clearly expresses that he is very satisfied, the problem is completely solved and he has positive feedback on the solution. The satisfaction score corresponds to that the problem is basically solved, but the customer may have made some suggestions for improvement or is not satisfied with some links. The low satisfaction score means that the customer feedback problem has not been effectively solved, or there are major objections to the solution, etc.;

[0062] Scoring implementation: Based on the above standards, each split historical problem solving record will be scored for satisfaction. If there is a satisfaction evaluation directly given by the customer in the record, it can be given priority reference. If not, a random selection will be conducted and then the customer service supervisor will make the scoring judgment;

[0063] Problem feature extraction: For each split problem description in the historical problem solving record, natural language processing technology is used to extract problem features that reflect the core characteristics of the problem;

[0064] Response extraction: Extract the response content given by customer service staff to the corresponding problem from the historical problem solving records to ensure that the extracted response is complete and accurately conveys the solution ideas at the time;

[0065] Judgment and saving operations: clarify the rules for quality analysis, that is, when the extracted problem features and the corresponding reply discourse are logically matched, and the customer satisfaction score corresponding to the record is high, the historical problem solving record is judged to be of high solution quality and can be saved to the enterprise management knowledge graph for subsequent reference.

[0066] S3. Combine the problem solving records with the problem characteristics and solution quality to generate a customer service capability profile for the assigned customer service. Then collect the customer's problem data and combine the problem data with the enterprise management knowledge graph to perform problem domain analysis. At the same time, perform problem characteristic analysis on the problem data.

[0067] The steps for S3 are as follows:

[0068] S3.1. Generate a customer service capability profile based on the problem solving records, problem characteristics, and solution quality of each customer service representative. Obtain the customer service capability profile of each customer service representative. The formula is as follows:

[0069]

[0070] Among them, H ij The number of high-quality problems solved by customer service in the jth problem area, Q ijk The resolution quality of the problem resolution record handled by the customer service can be high, medium, low, I(Q ijk = high) This is an indicator function. When it is equal to high, the function value is 1, indicating that the problem solving record is a high-quality solution; when it is not equal to high, the function value is 0, m ij is the problem solving record set of the customer service in the jth problem area, k is the counting index, when the customer service has m ij problem solving record, k is used to traverse this record;

[0071] T ij =m ij

[0072] Among them, T ij is the total number of problems handled by the customer service in the jth problem area;

[0073]

[0074] Among them, P ij represents the professional knowledge level of customer service in the jth problem area. ij =0, then P ij =0, which means that customer service has not handled any issues in this area. ij It reflects the level of application of customer service expertise in a specific field. The higher the value, the stronger the professional ability in that field.

[0075]

[0076] Among them, E i is the average problem solving efficiency index of customer service, t ijk The time spent by customer service to solve the corresponding problem, p is the total number of enterprise management problems divided into different fields, To calculate the total number of issues across all issue areas handled by customer service, First, sum up the resolution time corresponding to each problem area and each problem resolution record to get the total time it takes for customer service to resolve all problems;

[0077]

[0078] Among them, C i Used to measure the average communication ability of customer service, S ijk The satisfaction score given by customers to the communication with customer service in handling problems. The value range is determined according to the scoring rules set by the enterprise. Sum the customer service communication satisfaction scores across all problem areas and all problem resolution records;

[0079]

[0080] Among them, A i is the comprehensive ability score of customer service, which is used to evaluate the overall ability level of customer service. w1, w2, and w3 are the weights given by the enterprise to the three ability dimensions (professional knowledge ability, problem-solving efficiency, and communication ability) according to business needs, and w1+w2+w3=1. In order to comprehensively consider the professional knowledge and ability of customer service in different fields, the problem solving efficiency E i It is the average time spent on solving each problem. In order to make the efficiency higher (less time spent) and the comprehensive score higher, we take E i The inverse of is involved in the calculation. Through the above series of formulas, the ability of each customer service representative can be quantitatively analyzed from different dimensions, and then a more comprehensive and detailed customer service capability portrait can be generated.

[0081] S3.2. Collect customer problem data in real time from the enterprise management port, and combine the problem data with the enterprise management knowledge graph to perform problem domain analysis to obtain the problem domain corresponding to the problem data. At the same time, perform problem feature analysis on the problem data to obtain the problem features corresponding to the problem data.

[0082] Determine data collection channels: Identify various channels for customer feedback in enterprise management, and monitor these channels in real time through corresponding interface technologies or data collection tools to ensure that problem data submitted by customers can be obtained in the first place;

[0083] Combine knowledge graphs to analyze problem domains: call out the built enterprise management knowledge graph from the enterprise's knowledge graph storage system, then use natural language processing (NLP) technology to extract keywords from each problem data text collected in real time, and match the extracted keywords with the entities and relationships in the enterprise management knowledge graph. Consider the matching of each keyword comprehensively, and finally determine the problem domain to which each problem data belongs;

[0084] Problem feature analysis: Conduct a more in-depth semantic analysis of the problem data text. In addition to keyword extraction, natural language processing technologies such as syntactic analysis, named entity recognition, and topic models (such as LDA topic models) can also be used. Through these analyses, the deep semantic features hidden behind the problem can be excavated. For example, the focus of the problem is process optimization, resource allocation, personnel management, or other aspects; the subject scope involved in the problem is a single department or collaboration among multiple departments, etc. Then, based on the results of the above semantic analysis, a series of features that can comprehensively describe the problem are extracted to form a problem feature set.

[0085] S4. Perform similar matching on the problem features of the problem data and the problem features of the problem solving record, and display the corresponding reply utterance to the customer according to the matching result, and assign the customer to the corresponding customer service of the reply utterance;

[0086] The steps of S4 are as follows:

[0087] S4.1. Perform complexity and difficulty analysis on the problem features of the problem data, and dynamically set similarity thresholds based on the complexity and difficulty;

[0088] When the problem features are more complex and difficult, the similarity threshold is lower;

[0089] When the problem features are more complex and the difficulty is lower, the similarity threshold is higher. The specific formula is as follows:

[0090] Complexity(q i )=M1×N(q i )+M2×R PT (q i )+M3×L(q i )

[0091] Among them, Complexity(q i ) represents the problem data q i The comprehensive complexity and difficulty score of the problem features, N(q i ) is the problem data q i The number of problem features, R PT (q i ) is the problem data q i The professional term proportion index, L(qi ) is the problem data q i The logical correlation index between the problem features, M1, M2, M3 are the number of features, the proportion of professional terms, and the weights corresponding to the logical correlation;

[0092] T(q i )=T max -k T × Complexity(q i )

[0093] Among them, T(q i ) is the problem data q i Dynamically set similarity threshold, and 0≤T(q i )≤T max , T max is the upper limit of similarity threshold, k T is the adjustment factor.

[0094] S4.2, the problem features of the problem data are combined with the problem features of the problem solution record for similarity matching. When the similarity between the problem features of the problem data and the problem features of the problem solution record meets the similarity threshold, the reply speech corresponding to the problem features is displayed to the customer, and the customer is assigned to the corresponding customer service of the reply speech. On the contrary, when the similarity between the problem features of the problem data and the problem solution record does not meet the similarity threshold, enter S5, the formula is as follows:

[0095]

[0096] in, For the problem data q i Characteristics of the problem, Recording for problem solving j Problem characteristics, Sim(q i ,r j ) is the problem data q i Problem characteristics and problem solving records j The similarity of the problem features;

[0097] Sim(q i r j )≥T(q i ), the reply words that match the question characteristics are displayed to the customer, and the customer is assigned to the corresponding customer service of the reply words;

[0098] Sim(q i r j ) <T(q i ) and enter S5.

[0099] S4.2 also includes the following steps:

[0100] S4.2.1. When there are multiple problem features in the problem data, the problem features are compared in the problem domain. When all the problem features are in the same problem domain, the customer service with the highest similarity among all the problem feature matches is selected for assignment;

[0101] S4.2.2. When there are multiple problem features in the problem data, and the multiple problem features do not belong to the same problem field, the problem features are sorted according to the problem data, and the customer service with the highest similarity of the first problem feature is assigned.

[0102] S5. When S4 does not meet the matching requirements, the problem characteristics of the problem data are combined with the customer service capability profile and satisfaction indicators that match the problem field to conduct the best customer service analysis, and the customer is assigned to the corresponding customer service according to the analysis results.

[0103] The steps of S5 are as follows:

[0104] S5.1. Use recent response time, recent customer satisfaction score, and the number of people being received as satisfaction indicators;

[0105] From the customer service system database, find and extract the recent (one month) customer service staff response time data. The response time refers to the time interval from the customer asking a question to the customer service staff's first response.

[0106] Obtain recent customer satisfaction rating data for each customer service from the database. The satisfaction rating can be the customer's evaluation of the customer service after the problem is solved;

[0107] By real-time monitoring of customer service work status, the number of customers currently being received by each customer service representative can be counted. This can be achieved through real-time data from the customer service work platform or a dedicated monitoring system.

[0108] S5.2: When the similarity of all problem features in S4.2 does not meet the similarity threshold, the problem features of the problem data are combined with the customer service capability profile and satisfaction index that meet the problem domain to perform the best customer service analysis, and the customer is assigned to the corresponding customer service according to the analysis results. The specific steps are as follows:

[0109]

[0110] Among them, Score (S i ) is the comprehensive score of customer service satisfaction index, rt(S i ) is the response time, cs(S i ) is the satisfaction rating data, sr(S i) is the number of people being served. The inverse of the response time is taken here because the shorter the response time, the better. The higher the customer satisfaction score, the better. The fewer the number of people being served, the higher the service quality may be. Then the Score (S) of each customer service is calculated. i ) to find the maximum Score (S i ) to allocate the corresponding customer service.

[0111] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. A method for allocating customer service for enterprise management consulting based on problem characteristics, characterized in that: The following steps are involved: S1. Collect the historical problem solving records of customer service, and build the enterprise management knowledge map based on the historical problem solving records, and set the responsible areas of each customer service; S2. Analyze the problem solving quality based on historical problem solving records to obtain the problem solving quality of the problem solving records, and retain the problem features and corresponding response discourses of the problem solving records with high solution quality; S3. Combine the problem solving records with the problem characteristics and solution quality to generate a customer service capability profile for the assigned customer service. Then collect the customer's problem data and combine the problem data with the enterprise management knowledge graph to perform problem domain analysis. At the same time, perform problem feature analysis on the problem data. S4. Perform similar matching on the problem features of the problem data and the problem features of the problem solving record, and display the corresponding reply utterance to the customer according to the matching result, and assign the customer to the corresponding customer service of the reply utterance; S5. When S4 does not meet the matching requirements, the problem characteristics of the problem data are combined with the customer service capability profile and satisfaction indicators that match the problem field to conduct the best customer service analysis, and the customer is assigned to the corresponding customer service according to the analysis results.

2. The method for allocating enterprise management consulting customer service based on problem characteristics according to claim 1, characterized in that: The S1 is connected to the port managed by the enterprise, and then collects customer-related historical data from the port, and extracts historical problem solving records from the historical data.

3. The method for allocating enterprise management consulting customer service based on problem characteristics according to claim 1, characterized in that: The steps of S1 are as follows: S1.

1. Collect the business areas of the enterprise, and then combine the business areas of the enterprise with the historical problem solving records to construct the enterprise management knowledge graph in a structured manner, and use the enterprise management knowledge graph to determine the problem areas to which the historical problem solving records belong; S1.

2. Collect the list of customer service staff and historical problem solving records to determine the problem areas that each customer service representative has solved, and set them as the customer service representative’s area of ​​responsibility.

4. The method for allocating enterprise management consulting customer service based on problem characteristics according to claim 1, characterized in that: The steps of S2 are as follows: S2.

1. Split the overall historical problem solving records according to the problems and obtain multiple different historical problem solving records; S2.

2. Set up a satisfaction rating mechanism. The satisfaction rating is divided into high, medium, and low. At the same time, extract the problem features and corresponding reply discourses from multiple different historical problem solving records. Then, combine the problem features and corresponding reply discourses with the customer satisfaction ratings for quality analysis. When the problem features match the reply discourses and the customer satisfaction rating is high, the problem solving record is judged to be of high resolution quality, and the problem features and corresponding reply discourses are saved. Otherwise, they are not saved.

5. The method for allocating enterprise management consulting customer service based on problem characteristics according to claim 1, characterized in that: The steps of S3 are as follows: S3.

1. Generate a customer service capability profile based on the problem solving records, problem characteristics, and solution quality of each customer service representative, and obtain a customer service capability profile for each customer service representative; S3.

2. Collect customer problem data in real time from the enterprise management port, and combine the problem data with the enterprise management knowledge graph to perform problem domain analysis to obtain the problem domain corresponding to the problem data. At the same time, perform problem feature analysis on the problem data to obtain the problem features corresponding to the problem data.

6. The method for allocating enterprise management consulting customer service based on problem characteristics according to claim 1, characterized in that: The steps of S4 are as follows: S4.

1. Perform complexity and difficulty analysis on the problem features of the problem data, and dynamically set similarity thresholds based on the complexity and difficulty; When the problem features are more complex and difficult, the similarity threshold is lower; When the problem features are more complex and less difficult, the similarity threshold is higher; S4.

2. The problem features of the problem data are combined with the problem features of the problem solution record for similarity matching. When the similarity between the problem features of the problem data and the problem features of the problem solution record meets the similarity threshold, the reply utterance corresponding to the problem features is displayed to the customer, and the customer is assigned to the corresponding customer service of the reply utterance. Conversely, when the similarity between the problem features of the problem data and the problem features of the problem solution record does not meet the similarity threshold, enter S5.

7. The method for allocating enterprise management consulting customer service based on problem characteristics according to claim 6, characterized in that: The S4.2 further comprises the following steps: S4.2.

1. When there are multiple problem features in the problem data, the problem features are compared in the problem domain. When all the problem features are in the same problem domain, the customer service with the highest similarity among all the problem feature matches is selected for assignment; S4.2.

2. When there are multiple problem features in the problem data, and the multiple problem features do not belong to the same problem field, the problem features are sorted according to the problem data, and the customer service with the highest similarity of the first problem feature is assigned.

8. The method for allocating enterprise management consulting customer service based on problem characteristics according to claim 1, characterized in that: The steps of S5 are as follows: S5.

1. Use recent response time, recent customer satisfaction score, and the number of people being received as satisfaction indicators; S5.2: When the similarities of all problem features in S4.2 do not meet the similarity threshold, the problem features of the problem data are combined with the customer service capability profile and satisfaction index that match the problem domain to conduct the best customer service analysis, and the customer is assigned to the corresponding customer service according to the analysis results.

Images