Customer service problem processing method and device and storage medium

Abstract

The invention provides a customer service problem processing method and device and a storage medium, and the method comprises the steps: obtaining the problem information of an online customer service problem under the condition that a request for manually processing the online customer service problem is received, and extracting a first keyword from the problem information; based on the first keyword and a personnel ability scoring model, obtaining a processing ability score of the customer service personnel for processing the online customer service problem; determining a comprehensive score of the customer service staff according to the processing capability score of the customer service staff and the number of customer service problems currently processed by the customer service staff; and based on the comprehensive score and the number of the customer service problems, determining customer service personnel for processing the online customer service problems. According to the method and the device, intelligent routing distribution can be carried out on the online customer service problems, the online customer service problems can be quickly and accurately distributed to appropriate customer service personnel, and the processing efficiency and the service quality of the online customer service problems are improved.

Description

Technical Field

[0001] The present invention relates to the field of Internet technologies, and in particular, to a method and apparatus for processing customer service problems and a storage medium. Background Art

[0002] With the development and popularization of the Internet, more and more enterprises begin to pay attention to the customer experience, and improving customer satisfaction has become an important goal of enterprises. Customer service is an important channel for communication between enterprises and customers, and the quality and ability of customer service personnel directly affect customer satisfaction. Therefore, how to quickly and accurately allocate the needs of customers to appropriate customer service personnel has become a problem faced by enterprises. In practical applications, the processing capabilities of customer service personnel for different types of customer service problems are different. If a customer service problem is not allocated to an appropriate customer service personnel, it may cause the customer problem to not be processed in a timely manner, resulting in a decrease in customer satisfaction. Summary of the Invention

[0003] In view of this, a technical problem to be solved by the present invention is to provide a method and apparatus for processing customer service problems and a storage medium.

[0004] According to a first aspect of the present disclosure, there is provided a method for processing a customer service problem, including: when receiving a request for manually processing an online customer service problem, obtaining problem information of the online customer service problem, and extracting a first keyword from the problem information; obtaining a processing ability score of a customer service personnel for processing the online customer service problem based on the first keyword and a personnel ability scoring model; determining a comprehensive score of the customer service personnel according to the processing ability score of the customer service personnel and the number of customer service problems currently processed by the customer service personnel; determining a customer service personnel for processing the online customer service problem based on the comprehensive score and the number of customer service problems.

[0005] Optionally, the obtaining a processing ability score of a customer service personnel for processing the online customer service problem based on the first keyword and a personnel ability scoring model includes: determining a second keyword matching the first keyword in the personnel ability scoring model; obtaining the processing ability score according to a corresponding relationship between the second keyword and the processing ability score of the customer service personnel.

[0006] Optionally, constructing the personnel ability scoring model includes: obtaining an information interaction document of a customer service problem processed by the customer service personnel, and determining a second keyword corresponding to the information interaction document; determining a processing ability score of the customer service personnel for processing a customer service problem corresponding to the second keyword; constructing the personnel ability scoring model according to a corresponding relationship between the second keyword and the processing ability scores of each customer service personnel.

[0007] Optionally, determining the processing ability score of the customer service staff for handling the customer service issue corresponding to the second keyword includes: determining the weight score of the second keyword; obtaining the satisfaction and duration of the customer service staff for handling the customer service issue corresponding to the second keyword; and determining the processing ability score of the customer service staff for handling this customer service issue based on the weight score, the satisfaction, and the duration.

[0008] Optionally, determining the weight score of the second keyword includes: determining the number of occurrences of the second keyword in the information interaction document; determining the occurrence frequency of the second keyword based on the number of occurrences and the length information of the information interaction document; determining the inverse document frequency of the second keyword based on the number of information interaction documents containing the second keyword and the total number of information interaction documents; and determining the weight score of the second keyword based on the occurrence frequency, the inverse document frequency, the length of the information interaction document, the average length of all information interaction documents, and an adjustment factor.

[0009] Optionally, determining the second keyword corresponding to the information interaction document includes: performing word segmentation on the information interaction document to obtain a plurality of first word segments; determining the number of occurrences and weight information of each first word segment; determining the keyword score of each first word segment based on the number of occurrences, the weight information of each first word segment, a damping coefficient, and the total number of first word segments; and determining the second keyword based on the keyword scores of each first word segment.

[0010] Optionally, determining the second keyword based on the keyword scores of each first word segment includes: selecting at least one second word segment with the highest keyword score among the keyword scores of each first word segment as the second keyword.

[0011] Optionally, determining the comprehensive score of the customer service staff based on the processing ability score of the customer service staff and the number of customer service issues currently being handled by the customer service staff includes: obtaining the processing ability score of the customer service staff as the processing ability weight value; calculating the processing ability weight value ratio of the customer service staff based on the processing ability weight value and the number of customer service issues as the comprehensive score; and selecting the customer service staff for handling the online customer service issue among all customer service staff according to the processing ability weight value ratios of all customer service staff.

[0012] Optionally, selecting the customer service staff for handling the online customer service issue among all customer service staff according to the processing ability weight value ratios of all customer service staff includes: selecting the minimum processing ability weight value ratio among the processing ability weight value ratios of all customer service staff; and selecting the customer service staff for handling the online customer service issue based on the minimum processing ability weight value ratio.

[0013] Optionally, the selecting of the customer service staff for processing the online customer service problem based on the minimum processing capacity weight ratio includes: if the number of the minimum processing capacity weight ratios is 1, using the customer service staff corresponding to this minimum processing capacity weight ratio as the customer service staff for processing the online customer service problem; if the number of the minimum processing capacity weight ratios is multiple, selecting, from the multiple customer service staff corresponding to the multiple minimum processing capacity weight ratios, the customer service staff with the least number of currently processed customer service problems as the customer service staff for processing the online customer service problem.

[0014] Optionally, the extracting of the first keyword from the problem information includes: performing word segmentation on the problem information to obtain multiple second word segments; determining the occurrence times and weight information of each second word segment; determining the keyword score of each second word segment according to the occurrence times, weight information of each second word segment, the damping coefficient, and the total number of the second word segments; and determining the first keyword based on the keyword scores of each second word segment.

[0015] Optionally, the determining of the first keyword based on the keyword scores of each second word segment includes: selecting, from the keyword scores of each second word segment, the second word segment with the highest keyword score as the first keyword.

[0016] Optionally, when the processing of the online customer service problem is completed and the number of processed online customer service problems reaches a quantity threshold, an update process is performed on the personnel ability scoring model.

[0017] According to a second aspect of the present disclosure, there is provided a customer service problem processing apparatus, including: an information extraction module, configured to obtain problem information of the online customer service problem and extract a first keyword from the problem information when receiving a request for manually processing the online customer service problem; an ability scoring module, configured to obtain a processing ability score of a customer service staff for processing the online customer service problem based on the first keyword and a personnel ability scoring model; a comprehensive score module, configured to determine a comprehensive score of the customer service staff according to the processing ability score of the customer service staff and the number of currently processed customer service problems of the customer service staff; and a personnel allocation module, configured to determine the customer service staff for processing the online customer service problem based on the comprehensive score and the number of customer service problems.

[0018] According to a third aspect of the present disclosure, there is provided a customer service problem processing apparatus, including: a memory; and a processor coupled to the memory, where the processor is configured to execute the method as described above based on instructions stored in the memory.

[0019] According to a fourth aspect of the present disclosure, there is provided a computer-readable storage medium storing computer instructions that, when executed by a processor, perform the method as described above.

[0020] The customer service problem handling method, apparatus, and storage medium of the present disclosure extract keywords from online customer service problems, obtain the handling ability scores of customer service personnel based on the keywords and the personnel ability scoring model, and determine the customer service personnel for handling online customer service problems according to the handling ability scores and the number of current online customer service problems being handled; it can perform intelligent routing and distribution of online customer service problems, quickly and accurately assign online customer service problems to appropriate customer service personnel, improve the accuracy of classification and assignment of online customer service problems, and improve the handling efficiency and service quality of online customer service problems. Description of the Drawings

[0021] By describing the embodiments of the present disclosure in more detail in conjunction with the drawings, the above and other objects, features, and advantages of the present disclosure will become more apparent. The drawings are used to provide a further understanding of the embodiments of the present disclosure, and constitute a part of the specification. Together with the embodiments of the present disclosure, they are used to explain the present disclosure and do not constitute a limitation to the present disclosure. The following describes the above and other objects and advantages of the present disclosure in more detail with reference to specific embodiments and the drawings. In the drawings, the same or corresponding technical features or components will be represented by the same or corresponding reference numerals.

[0022] Figure 1 It is a flowchart showing an embodiment of the customer service problem handling method according to the present disclosure;

[0023] Figure 2 It is a flowchart showing the process of determining the handling ability score of a customer service personnel in an embodiment of the customer service problem handling method according to the present disclosure;

[0024] Figure 3 It is a flowchart showing the process of constructing a personnel ability scoring model in an embodiment of the customer service problem handling method according to the present disclosure;

[0025] Figure 4 It is a flowchart showing the process of determining the weight score in an embodiment of the customer service problem handling method according to the present disclosure;

[0026] Figure 5 It is a schematic diagram of a personnel ability scoring model in an embodiment of the customer service problem handling method according to the present disclosure;

[0027] Figure 6 It is a flowchart showing the process of determining the customer service personnel for handling online customer service problems in an embodiment of the customer service problem handling method according to the present disclosure;

[0028] Figure 7 Schematic diagram of a module according to an embodiment of a customer service problem processing device of the present disclosure;

[0029] Figure 8 Schematic diagram of a module according to another embodiment of a customer service problem processing device of the present disclosure;

[0030] Figure 9 Schematic diagram of a model construction module in an embodiment of a customer service problem processing device of the present disclosure;

[0031] Figure 10 Schematic diagram of a module according to yet another embodiment of a customer service problem processing device of the present disclosure. Detailed implementation manners

[0032] Hereinafter, exemplary embodiments of the present disclosure will be described in conjunction with the accompanying drawings. For the sake of clarity and conciseness, not all features of the embodiments are described in the specification. However, it should be understood that many specific settings specific to the implementation manner must be made during the implementation of the embodiments in order to achieve the specific goals of the developers. For example, those restrictions related to the equipment and business are met, and these restrictions may vary with different implementation manners. In addition, it should also be understood that although the development work may be very complex and time-consuming, for those skilled in the art who benefit from the content of the present disclosure, such development work is only a routine task.

[0033] It should be noted that: Unless otherwise specifically stated, the relative arrangements, numerical expressions and values of the components and steps set forth in these embodiments do not limit the scope of the present disclosure.

[0034] Those skilled in the art can understand that the terms "first", "second", etc. in the embodiments of the present disclosure are only used to distinguish different steps, devices or modules, etc., and neither represent any specific technical meaning nor indicate an inevitable logical order between them.

[0035] It should also be understood that in the embodiments of the present disclosure, "a plurality of" may refer to two or more, and "at least one" may refer to one, two or more.

[0036] It should also be understood that for any component, data or structure mentioned in the embodiments of the present disclosure, without clear limitation or contrary indication in the context, it can generally be understood as one or more.

[0037] In addition, the term "and / or" in the present disclosure merely describes the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in the present disclosure generally represents an "or" relationship between the associated objects before and after.

[0038] It should also be understood that the description of each embodiment in the present disclosure emphasizes the differences between the embodiments. The same or similar parts can be referred to each other. For the sake of brevity, they will not be described one by one.

[0039] At the same time, it should be understood that for the convenience of description, the sizes of the various parts shown in the drawings are not drawn in actual proportional relationships.

[0040] The following description of at least one exemplary embodiment is actually merely illustrative and in no way restricts the present disclosure or its application or use.

[0041] For technologies, methods, and devices known to those of ordinary skill in the relevant art, they may not be discussed in detail, but where appropriate, the said technologies, methods, and devices should be regarded as part of the specification.

[0042] It should be noted that similar reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0043] In addition, to avoid obscuring the present disclosure with unnecessary details, only the processing steps and / or device structures closely related to the solution according to the present disclosure are shown in the drawings, while other details less related to the present disclosure are omitted. It should also be noted that similar reference numerals and letters in the drawings indicate similar items, and thus once an item is defined in one drawing, it does not need to be further discussed for subsequent drawings.

[0044] Figure 1 It is a schematic flowchart of an embodiment of the customer service problem processing method according to the present disclosure, as Figure 1 shown:

[0045] Step 101, when receiving a request for manually processing an online customer service problem, obtain the problem information of the online customer service problem and extract the first keyword from the problem information.

[0046] In one embodiment, the online customer service questions raised by customers can be automatically processed by a customer service robot or manually processed by customer service staff. Customers can raise online customer service questions through web pages, phones, APPs, etc. Customers can choose to have the online customer service questions processed manually or automatically by the customer service robot. The online customer service questions can be various types of questions, such as product after-sales, consultation, complaint and other questions.

[0047] When receiving a request from a customer to manually process an online customer service question, obtain the question information input by the customer through the web page or APP. The question information can include text information such as question type and question description; it is also possible to parse the voice input by the customer through the phone to obtain question information including information such as question type and question description; various methods can be used to extract the first keyword from the question information.

[0048] Step 102, based on the first keyword and the personnel ability scoring model, obtain the processing ability score of the customer service staff for processing the online customer service question.

[0049] Step 103, according to the processing ability score of the customer service staff and the number of customer service questions currently being processed by the customer service staff, determine the comprehensive score of the customer service staff.

[0050] Step 104, based on the comprehensive score and the number of customer service questions, determine the customer service staff for processing the online customer service question.

[0051] The customer service question processing method in the above embodiment extracts the first keyword from the question information of the online customer service question, obtains the processing ability score of the customer service staff based on the first keyword and the personnel ability scoring model, and determines the customer service staff for processing the online customer service question according to the processing ability score and the number of customer service questions currently being processed; it can perform intelligent routing and distribution of customer service questions, and can quickly and accurately assign customer service questions to appropriate customer service staff, improving customer satisfaction.

[0052] Figure 2 As shown in the following figure, it is a flow chart of determining the processing ability score of the customer service staff in an embodiment of the customer service question processing method according to the present disclosure. Figure 2 As shown below:

[0053] Step 201, in the personnel ability scoring model, determine the second keyword that matches the first keyword.

[0054] Step 202, according to the corresponding relationship between the second keyword and the processing ability score of the customer service staff, obtain the processing ability score.

[0055] In one embodiment, a personnel ability scoring model is pre-constructed, and a corresponding relationship between a second keyword and the processing ability score of a customer service staff is pre-set in the personnel ability scoring model. By querying in the personnel ability scoring model, the first keyword is compared with the second keyword in the personnel ability scoring model to determine the second keyword that matches the first keyword; in the personnel ability scoring model, the corresponding relationship between this second keyword and the processing ability score of the customer service staff is obtained, and according to this corresponding relationship, the processing ability score of the customer service staff for handling the online customer service problem raised by the current customer is obtained.

[0056] The personnel ability scoring model can be various models, and the personnel ability scoring model can adopt various model construction methods. Figure 3 FIG. is a schematic flowchart of constructing a personnel ability scoring model in an embodiment of the customer service problem processing method according to the present disclosure, as Figure 3 shown:

[0057] Step 301: Obtain the information interaction document of the customer service problem handled by the customer service staff, and determine the second keyword corresponding to the information interaction document.

[0058] In one embodiment, the information interaction documents of the customer service problems handled by each customer service staff can be obtained. The information interaction document can be a record document of the information interaction between the customer service staff and the customer through the web page, APP when the customer service staff solves the customer service problem for the customer, and a dialogue document of the conversation between the customer service staff and the customer through the phone, etc.

[0059] Step 302: Determine the processing ability score of the customer service staff for handling the customer service problem corresponding to the second keyword.

[0060] Step 303: Construct a personnel ability scoring model according to the corresponding relationship between the second keyword and the processing ability scores of each customer service staff.

[0061] In one embodiment, multiple methods can be used to determine the second keyword corresponding to the information interaction document. Perform word segmentation processing on the information interaction document to obtain multiple first word segments; determine the occurrence times and weight information of each first word segment; according to the occurrence times, weight information of each first word segment, as well as the damping coefficient and the total number of first word segments, determine the keyword score of each first word segment; based on the keyword scores of each first word segment, determine the second keyword. At least one word segment with the highest keyword score can be selected from the keyword scores of each first word segment as the second keyword.

[0062] For example, multiple information exchange documents of customer service issues handled by various customer service personnel are obtained, and multiple existing word segmentation methods can be used to perform word segmentation processing on the entire information exchange document or a part of the information exchange document (issue type and / or issue description information) to obtain multiple first word segmentations. The word segmentation method can be a two-way maximum matching algorithm, etc.

[0063] The information in an information exchange document includes the problem description information "I want to complain, complain that your customer service attitude has problems". "I want to complain, complain that your customer service attitude has problems" is segmented, and multiple segmented words are obtained: "I", "want", "complain", "complaint", "you", "of", "customer service", "attitude", "have", "problem", etc. Preprocess multiple segmented words, including removing stop words, punctuation marks and numbers, and converting the text to lowercase. The first segmented word is: "want", "complain", "complaint", "customer service", "attitude", "problem".

[0064] The number of occurrences (frequency) of each first participle is determined as follows: the number of occurrences (frequency) of "want" is 1; the number of occurrences (frequency) of "complaint" is 2; the number of occurrences (frequency) of "customer service" is 1; the number of occurrences (frequency) of "attitude" is 1; and the number of occurrences (frequency) of "problem" is 1.

[0065] Weight information is set for each first participle, and the weight information can be determined by using a variety of existing methods, and the weight information can be determined according to specific needs, frequency of occurrence, context, etc. For example, the weight information is predetermined as follows: "w1": 0.4, "w2": 0.6, and the weight information of "want" is set to 0.4, the weight information of "complaint" is set to 0.6, the weight information of "customer service" is set to 0.4, the weight information of "attitude" is set to 0.4, and the weight information of "problem" is set to 0.4 according to the frequency of occurrence of the first participle.

[0066] Based on the following formula (1-1), the keyword score of each first participle is determined according to the number of occurrences of each first participle, the weight information, the damping coefficient and the total number of first participles:

[0067] Rake=(1-d)×(wi×freqi) / (d+totalWords)(1-1);

[0068] Among them, Rake is the keyword score, d is the damping coefficient (representing the threshold of the distance between keywords), wi represents the weight information of the i-th word segmentation, freqi represents the number of occurrences of the i-th word segmentation (representing the frequency of occurrence of the word segmentation in the information interaction document), and totalWords represents the total number of word segmentations.

[0069] Assume that the damping coefficient d is 0.85, determine that the total number of the first participles is 5, and calculate the keyword scores Rake of each participle according to formula (1-1):

[0070] The keyword score of "want": (1 - 0.85)×(0.4×1) / (0.85 + 5) ≈ 0.013;

[0071] The keyword score of "complain": (1 - 0.85)×(0.6×2) / (0.85 + 5) ≈ 0.015;

[0072] The keyword score of "customer service": (1 - 0.85)×(0.4×1) / (0.85 + 5) ≈ 0.013;

[0073] The keyword score of "attitude": (1 - 0.85)×(0.4×1) / (0.85 + 5) ≈ 0.013;

[0074] The keyword score of "problem": (1 - 0.85)×(0.4×1) / (0.85 + 5) ≈ 0.013.

[0075] Among the keyword scores of each first participle, select at least one participle with the highest keyword score as the second keyword, that is, determine "complain" as the second keyword. If the number of participles with the highest keyword score is multiple, one participle can be selected from the multiple participles with the highest keyword score according to factors such as the occurrence frequency as the second keyword.

[0076] In one embodiment, multiple methods can be used to extract the first keyword from the problem information of the online customer service problem, and the same method as determining the second keyword can be used. For example, perform participle processing on the problem information to obtain multiple second participles; determine the occurrence times and weight information of each second participle; according to the occurrence times and weight information of each second participle, as well as the damping coefficient and the total number of second participles, determine the keyword scores of each second participle; based on the keyword scores of each second participle, determine the first keyword. One participle with the highest keyword score can be selected from the keyword scores of each second participle as the first keyword.

[0077] For example, the problem information of an online customer service issue is "I want to file a complaint about the product quality". After performing word segmentation on "I want to file a complaint about the product quality", multiple segmented words obtained are: "I", "want", "to file a complaint", "to file a complaint", "product", "quality", etc. After preprocessing the multiple segmented words, the second segmented words obtained are: "want", "to file a complaint", "to file a complaint", "product", "quality". Calculate the occurrence frequency of each second segmented word: the occurrence times (frequency) of "want" is 1; the occurrence times (frequency) of "to file a complaint" is 2; the occurrence times (frequency) of "product" is 1; the occurrence times (frequency) of "quality" is 1.

[0078] Assign weight information to each second segmented word. According to the occurrence frequency of the second segmented word, set the weight information of "want" to 0.4, the weight information of "to file a complaint" to 0.6, the weight information of "product" to 0.4, and the weight information of "quality" to 0.4. Assume that the damping coefficient d is 0.85, and determine that the total number of the first segmented words is 4. The keyword score Rake of each segmented word can be calculated according to formula (1-1):

[0079] The keyword score of "want": (1 - 0.85) × (0.4 × 1) / (0.85 + 4) ≈ 0.012;

[0080] The keyword score of "to file a complaint": (1 - 0.85) × (0.6 × 2) / (0.85 + 4) ≈ 0.037;

[0081] The keyword score of "product": (1 - 0.85) × (0.4 × 1) / (0.85 + 4) ≈ 0.012;

[0082] The keyword score of "quality": (1 - 0.85) × (0.4 × 1) / (0.85 + 4) ≈ 0.012;

[0083] Among the keyword scores of each second segmented word, select at least one segmented word with the highest keyword score as the first keyword, that is, determine "to file a complaint" as the first keyword. If the number of segmented words with the highest keyword score is multiple, one segmented word can be selected from the multiple segmented words with the highest keyword score according to factors such as occurrence frequency as the first keyword.

[0084] In one embodiment, multiple methods can be used to determine the processing ability score of a customer service staff in handling a customer service issue corresponding to a second keyword. For example, determine the weight score of the second keyword, and multiple methods can be used to determine the weight score; obtain the satisfaction and duration of the customer service staff in handling the customer service issue corresponding to the second keyword; determine the processing ability score of the customer service staff in handling this customer service issue according to the weight score, satisfaction, and duration.

[0085] Figure 4Schematic diagram of the process for determining the weight score in an embodiment of the customer service problem handling method according to the present disclosure, as shown in Figure 4 shown below:

[0086] Step 401: Determine the number of occurrences of the second keyword in the information interaction document.

[0087] Step 402: Determine the occurrence frequency of the second keyword according to the number of occurrences and the length information of the information interaction document.

[0088] Step 403: Determine the inverse document frequency of the second keyword based on the number of information interaction documents containing the second keyword and the total number of information interaction documents.

[0089] Step 404: Determine the weight score of the second keyword according to the occurrence frequency, inverse document frequency, length of the information interaction document, average length of all information interaction documents, and adjustment factor.

[0090] For example, the information interaction document for the customer service problem handled by the customer service staff is the dialogue document between the customer service staff and the customer, and the number of dialogue documents is five. The second keyword corresponding to a dialogue passage is determined to be "complaint", and the number of occurrences of "complaint" in the five dialogue documents is determined. According to the number of occurrences and the length information of the dialogue document, the occurrence frequency of "complaint" is as follows:

[0091] For dialogue document 1: "Complaint" appears 10 times, and the document length of dialogue document 1 is 100, so the occurrence frequency TF1 of "complaint" = 10 / 100 = 0.1;

[0092] For dialogue document 2: "Complaint" appears 20 times, and the document length of dialogue document 2 is 200, so the occurrence frequency TF2 of "complaint" = 20 / 200 = 0.1;

[0093] For dialogue document 3: "Complaint" appears 0 times, and the document length of dialogue document 3 is 100, so the occurrence frequency TF3 of "complaint" = 0;

[0094] For dialogue document 4: "Complaint" appears 30 times, and the document length of dialogue document 4 is 300, so the occurrence frequency TF4 of "complaint" = 30 / 300 = 0.1;

[0095] For dialogue document 5: "Complaint" appears 0 times, and the document length of dialogue document 5 is 100, so the occurrence frequency TF5 of "complaint" = 0.

[0096] Existing methods can be used to determine the inverse document frequency of the second keyword based on the number of information interaction documents containing the second keyword and the total number of information interaction documents. For example, the TF of "complaint" (i.e., the number of conversation documents containing "complaint") = 3, and the total number of conversation documents N = 5. Therefore, the inverse document frequency IDF of "complaint" = log(5 / 3).

[0097] The following formula (1-2) can be used to determine the weight score of the second keyword based on the occurrence frequency, inverse document frequency, length of the information interaction document, average length of all information interaction documents, and adjustment factors:

[0098] CM20 = IDF × (TF * (k1 + 1)) / (TF + k1 * (1 - b + b * (D / avgdl))) (1-2);

[0099] Where CM20 is the weight score of the keyword; TF represents the occurrence frequency of the keyword in the information interaction document, IDF represents the inverse document frequency of the keyword, D represents the document length of the information interaction document, avgdl represents the average document length of the information interaction document, and k1 and b are adjustment factors (k1 and b can represent the urgency level, etc., and can be configured according to the actual situation).

[0100] For example, assuming k1 = 2, b = 0.75, and the average document length avgdl = 180 ((100 + 200 + 100 + 300 + 100) / 5), the CM20 score for calculating the weight score of "complaint" based on formula (1-2) is: CM20 = log(5 / 3) × (0.1 * (2 + 1)) / (0.1 + 2 * (1 - 0.75 + 0.75 * (100 / 180))). In practical applications, each conversation document may contain many keywords, and a score needs to be calculated for each keyword.

[0101] To obtain the satisfaction and duration of the customer service staff in handling the customer service problem corresponding to the second keyword, the following formula (1-3) can be used to determine the handling ability score of the customer service staff in handling this customer service problem based on the weight score, satisfaction, and duration:

[0102] SL20 = CM20 × (SA / 10) × ST (1-3);

[0103] Where CM20 is the weight score of the second keyword, SA represents the satisfaction of the customer service staff in handling the problem corresponding to the second keyword (representing the satisfaction of the customer with the solution of the problem by the customer service staff), and ST represents the duration of the customer service staff in handling the problem corresponding to the second keyword.

[0104] For example, customer service staff A handled a problem with the second keyword "complaint" and calculated the weight score CM20 of the second keyword "complaint" to be 2.5 based on formula (1-2). The satisfaction SA of customer service staff A in handling the problem corresponding to the second keyword can be obtained through methods such as questionnaires and online evaluations. For example, in a 10-point evaluation system, the customer gave an 8-point satisfaction for the problem with the second keyword "complaint". The duration (ST) of customer service staff A in handling this problem can be recorded from actual operations. For example, it took customer service staff A 5 minutes to solve this problem.

[0105] Based on formula (1-3), the processing ability score of customer service staff A in handling the customer service problem corresponding to the second keyword "complaint" can be calculated as: SL20 = 2.5×(8 / 10)×5; that is, the processing ability score of customer service staff A in handling this is 10.

[0106] In one embodiment, the constructed personnel ability scoring model is as Figure 5 shown. The second keywords are "information security", "complaint", "loan", "contract", "card binding", etc. It is possible to determine the processing ability scores of each customer service staff A, customer service staff B, etc. in handling the customer service problems corresponding to the second keywords "information security", "complaint", "loan", "contract", "card binding", etc., and establish the corresponding relationship between the second keywords "information security", "complaint", "loan", "contract", "card binding" and the processing ability scores of each customer service staff. For example, the processing ability scores of customer service staff A in handling the customer service problems corresponding to the second keywords "information security", "complaint", "loan", "contract", "card binding", etc. are 20, 25, 20, 0, 10, etc.

[0107] Figure 6 FIG. is a schematic flow chart for determining the customer service staff handling the online customer service problem in an embodiment of the customer service problem handling method according to the present disclosure, as Figure 6 shown:

[0108] Step 601, obtain the processing ability score of the customer service staff as the processing ability weight value.

[0109] Step 602, based on the processing ability weight value and the number of customer service problems currently handled by the customer service staff, calculate the processing ability weight value ratio of the customer service staff as the comprehensive score.

[0110] Step 603, select the customer service staff handling the online customer service problem from all the customer service staff according to the processing ability weight value ratios of all the customer service staff.

[0111] In one embodiment, among the processing capacity weight ratios of all customer service staff, the minimum processing capacity weight ratio is selected. Based on the minimum processing capacity weight ratio, the customer service staff for handling online customer service issues can be selected. If the number of the minimum processing capacity weight ratios is 1, the customer service staff corresponding to this minimum processing capacity weight ratio is used as the customer service staff for handling online customer service issues; if the number of the minimum processing capacity weight ratios is multiple, among the multiple customer service staff corresponding to the multiple minimum processing capacity weight ratios, the customer service staff with the least number of currently handled customer service issues is selected as the customer service staff for handling online customer service issues.

[0112] For example, based on the personnel ability scoring model, the processing ability scores of each customer service staff for online customer service issues are obtained as the processing capacity weights wi, where wi represents the processing ability of the i-th customer service staff. The number of currently handled customer service issues ci of each customer service staff is obtained, where ci represents the number of currently handled customer service issues of the i-th customer service staff.

[0113] Using the following formula (1-4), based on the processing capacity weight and the number of customer service issues, the processing capacity weight ratio of the customer service staff is calculated as

[0114] Qi = (cj + 1) / wi (1-4);

[0115] where Qi is the processing capacity weight ratio (comprehensive score) of the i-th customer service staff and can be used as the comprehensive score of the i-th customer service staff; wi represents the processing capacity weight of the i-th customer service staff; ci represents the number of currently handled customer service issues of the i-th customer service staff.

[0116] Suppose there are currently three customer service staff, namely customer service staff A, customer service staff B, and customer service staff C. The first keyword extracted from the problem information of the online customer service issue is "verification code". Based on the "verification code" in the personnel ability scoring model, the second keyword "verification code" that matches the first keyword is determined. According to the corresponding relationship between the second keyword "verification code" and the processing ability scores of customer service staff A, customer service staff B, and customer service staff C, the processing ability scores (SL20 scores) of customer service staff A, customer service staff B, and customer service staff C are obtained, and the number of currently handled customer service issues of customer service staff A, customer service staff B, and customer service staff C is obtained.

[0117] For example, for customer service staff A: the processing ability score wi = 2, and the number of currently handled customer service issues ci = 1; for customer service staff B: the processing ability score wi = 3, and the number of currently handled customer service issues ci = 2; for customer service staff C: the processing ability score wi = 1, and the number of currently handled customer service issues ci = 0.

[0118] Calculate the processing ability weight ratios of customer service representatives A, B, and C based on formula (1-4) as the comprehensive score: The processing ability weight ratio of customer service representative A: (1 + 1) / 2 = 1; The processing ability weight ratio of customer service representative B: (2 + 1) / 3 = 1; The processing ability weight ratio of customer service representative C: (0 + 1) / 1 = 1.

[0119] If there are multiple minimum processing ability weight ratios, select the customer service representative with the fewest current number of processed issues. For example, the processing ability weight ratios of customer service representatives A, B, and C (which are all the same) are used as the three minimum processing ability weight ratios; among the three customer service representatives corresponding to the three minimum processing ability weight ratios, select customer service representative C with the fewest current number of processed customer service issues as the customer service representative for handling online customer service issues, assign this online customer service issue to customer service representative C, and increment the current number of processed issues of customer service representative C by 1. Use the above method until all online customer service issues that require manual processing are completed.

[0120] The customer service issue handling method in the above embodiment can, based on the processing ability of customer service representatives and their current workload, as fairly as possible allocate the issue response work to customer service representatives; in actual applications, the processing ability and current workload of customer service representatives can be adjusted according to actual situations to obtain a more accurate allocation result.

[0121] When the online customer service issue handling is completed, perform an update process on the personnel ability scoring model. For example, when the online customer service issue handling is completed, or when the number of processed online customer service issues reaches the quantity threshold, re-obtain the information interaction document of the customer service issues processed by the customer service representatives and determine the second keyword corresponding to the information interaction document; determine the processing ability score of the customer service representatives for handling the customer service issues corresponding to the second keyword; update the personnel ability scoring model according to the correspondence between the second keyword and the processing ability scores of each customer service representative. As the historical samples increase, the personnel ability scoring model can be continuously improved and updated; based on historical data analysis, the existing personnel ability scoring model can be improved, and as the number of customer consultations and the types of issues change, the personnel ability scoring model can be adaptively updated and improved, continuously improving the classification accuracy of customer service issues and the issue handling efficiency.

[0122] The customer service issue handling method in the above embodiment can perform intelligent routing and distribution of online customer service issues, improving the processing efficiency and service quality of online customer service issues, and enhancing customer satisfaction and usage experience.

[0123] In one embodiment, as Figure 7As shown in the figure, the present disclosure provides a customer service problem processing device 70, including an information extraction module 71, an ability scoring module 72, a comprehensive score module 73, and a personnel allocation module 74. When the information extraction module 71 receives a request for manually processing an online customer service problem, it obtains the problem information of the online customer service problem and extracts the first keyword from the problem information.

[0124] The ability scoring module 72 obtains the processing ability score of the customer service staff for processing the online customer service problem based on the first keyword and the personnel ability scoring model. The comprehensive score module 73 determines the comprehensive score of the customer service staff according to the processing ability score of the customer service staff and the number of customer service problems currently processed by the customer service staff. The personnel allocation module 74 determines the customer service staff for processing the online customer service problem based on the comprehensive score and the number of customer service problems.

[0125] In one embodiment, the information extraction module 71 performs word segmentation processing on the problem information to obtain a plurality of second word segments; the information extraction module 71 determines the occurrence times and weight information of each second word segment; the information extraction module 71 determines the keyword score of each second word segment according to the occurrence times, weight information of each second word segment, the damping coefficient, and the total number of second word segments; the information extraction module 71 determines the first keyword based on the keyword scores of each second word segment. For example, the information extraction module 71 selects the second word segment with the highest keyword score among the keyword scores of each second word segment as the first keyword.

[0126] The ability scoring module 72 determines the second keyword that matches the first keyword in the personnel ability scoring model, and the ability scoring module 72 obtains the processing ability score according to the corresponding relationship between the second keyword and the processing ability score of the customer service staff.

[0127] The personnel allocation module 73 obtains the processing ability score of the customer service staff as the processing ability weight value; the personnel allocation module 73 calculates the processing ability weight value ratio of the customer service staff based on the processing ability weight value and the number of customer service problems as the comprehensive score; the personnel allocation module 73 selects the customer service staff for processing the online customer service problem from all the customer service staff according to the processing ability weight value ratios of all the customer service staff.

[0128] The personnel allocation module 73 may select the minimum processing capacity weight ratio among the processing capacity weight ratios of all customer service personnel, and select the customer service personnel for handling online customer service problems based on the minimum processing capacity weight ratio. For example, if the number of the minimum processing capacity weight ratios is 1, the personnel allocation module 73 will use the customer service personnel corresponding to this minimum processing capacity weight ratio as the customer service personnel for handling online customer service problems; if the number of the minimum processing capacity weight ratios is multiple, the personnel allocation module 73 will select the customer service personnel with the least number of currently handled customer service problems among the multiple customer service personnel corresponding to the multiple minimum processing capacity weight ratios as the customer service personnel for handling online customer service problems.

[0129] In one embodiment, as Figure 8 shown, the customer service problem processing device 70 further includes a model construction module 75 and a model update module 76. The model construction module 75 obtains the information interaction document of the customer service problems handled by the customer service personnel, and determines the second keyword corresponding to the information interaction document; the model construction module 75 determines the processing capacity score of the customer service personnel for handling the customer service problems corresponding to the second keyword; the model construction module 75 constructs a personnel capacity score model according to the corresponding relationship between the second keyword and the processing capacity scores of each customer service personnel. The model update module 76 performs an update process on the personnel capacity score model when the online customer service problem is processed.

[0130] In one embodiment, as Figure 9 shown, the model construction module 75 includes a scoring unit 751, a weight determination unit 752, and a keyword determination unit 753. The scoring unit 751 determines the weight score of the second keyword, and obtains the satisfaction degree and duration of the customer service personnel for handling the customer service problems corresponding to the second keyword; the scoring unit 751 determines the processing capacity score of the customer service personnel for handling this customer service problem according to the weight score, the satisfaction degree, and the duration.

[0131] The weight determination unit 752 determines the number of occurrences of the second keyword in the information interaction document, and determines the occurrence frequency of the second keyword according to the number of occurrences and the length information of the information interaction document; the weight determination unit 752 determines the inverse document frequency of the second keyword based on the number of information interaction documents containing the second keyword and the total number of information interaction documents; the weight determination unit 752 determines the weight score of the second keyword according to the occurrence frequency, the inverse document frequency, the length of the information interaction document, the average length of all information interaction documents, and the adjustment factor.

[0132] The keyword determination unit 753 performs word segmentation on the information interaction document to obtain a plurality of first segmented words; the keyword determination unit 753 determines the occurrence times and weight information of each first segmented word; according to the occurrence times, weight information of each first segmented word, as well as the damping coefficient and the total number of first segmented words, it determines the keyword scores of each first segmented word; the keyword determination unit 753 determines the second keywords based on the keyword scores of each first segmented word. For example, the keyword determination unit 753 selects at least one segmented word with the highest keyword score among the keyword scores of each first segmented word as the second keyword.

[0133] In one embodiment, as Figure 10 shown, the present disclosure provides a customer service problem processing device, which may include a memory 82, a processor 81, a communication interface 83, and a bus 84. The memory 82 is used to store instructions, the processor 81 is coupled to the memory 82, and the processor 81 is configured to execute the above-mentioned customer service problem processing method based on the instructions stored in the memory 82.

[0134] The memory 82 may be a high-speed RAM memory, a non-volatile memory, etc., and the memory 82 may also be a memory array. The memory 82 may also be partitioned, and the partitions may be combined into virtual volumes according to certain rules. The processor 81 may be a central processing unit CPU, or an application specific integrated circuit ASIC (Application Specific Integrated Circuit), or one or more integrated circuits configured to implement the customer service problem processing method of the present disclosure.

[0135] In one embodiment, the present disclosure provides a computer-readable storage medium storing computer instructions, and when the instructions are executed by a processor, the method in any of the above embodiments is implemented.

[0136] The computer-readable storage medium may adopt any combination of one or more readable media. The readable media may be a readable signal medium or a readable storage medium. The readable storage medium may, for example, include but is not limited to an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples (non-exhaustive list) of the readable storage medium may include: an electrical connection having one or more wires, a portable disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above.

[0137] An embodiment of the present disclosure may also be a computer program product, which includes computer program instructions that, when run on a processor, cause the processor to execute the steps in the methods according to various embodiments of the present disclosure described in the "Exemplary Methods" section above in this specification.

[0138] For the customer service problem handling method, device, and storage medium in the above embodiments, keywords are extracted from the problem information of the online customer service problem, and based on the keywords and the personnel ability scoring model, the processing ability score of the customer service personnel is obtained. According to the processing ability score and the number of current customer service problems being processed, the customer service personnel for handling the online customer service problem can be determined; the online customer service problem can be intelligently routed and distributed, and the online customer service problem can be quickly and accurately assigned to the appropriate customer service personnel, improving the processing efficiency and service quality of the online customer service problem, and improving the customer satisfaction and usage experience.

[0139] The basic principles of the present disclosure have been described above in conjunction with specific embodiments. However, it should be noted that the advantages, benefits, effects, etc. mentioned in the present disclosure are only examples and not limitations, and it cannot be considered that these advantages, benefits, effects, etc. are essential for each embodiment of the present disclosure. In addition, the above-mentioned specific details of the disclosure are only for the purposes of illustration and easy understanding, rather than limitations, and the above details do not limit the present disclosure to necessarily adopt the above specific details to implement.

[0140] Each embodiment in this specification is described in a progressive manner. The key point of each embodiment is to illustrate the differences from other embodiments. For the same or similar parts between each embodiment, reference can be made to each other. For the system embodiment, since it basically corresponds to the method embodiment, the description is relatively simple, and for the relevant parts, reference can be made to the partial description of the method embodiment.

[0141] The block diagrams of the devices, apparatuses, equipment, and systems involved in the present disclosure are only illustrative examples and do not intend to require or imply that they must be connected, arranged, and configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, equipment, and systems can be connected, arranged, and configured in any manner. Words such as "including", "comprising", "having", etc. are open-ended words, meaning "including but not limited to", and can be used interchangeably with each other. The words "or" and "and" used here refer to the word "and / or", and can be used interchangeably with each other unless the context clearly indicates otherwise. The word "such as" used here refers to the phrase "such as but not limited to", and can be used interchangeably with each other.

[0142] It should also be noted that in the devices, equipment, and methods of the present disclosure, each component or each step can be decomposed and / or recombined. These decompositions and / or recombinations should be regarded as equivalent solutions of the present disclosure.

[0143] The foregoing description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present disclosure. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present disclosure. Thus, the present disclosure is not intended to be limited to the aspects shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

[0144] The foregoing description has been presented for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the present disclosure to the form disclosed herein. Although several example aspects and embodiments have been discussed above, those skilled in the art will appreciate that the foregoing embodiments are illustrative only and do not limit the scope of the present disclosure. Those skilled in the art should understand that the foregoing embodiments may be combined, modified, or substituted without departing from the scope and spirit of the present disclosure.

Claims

1. A customer service problem handling method, comprising: When receiving a request for manually handling an online customer service problem, obtaining the problem information of the online customer service problem, and extracting a first keyword from the problem information; Based on the first keyword and a personnel ability scoring model, obtaining a handling ability score of a customer service personnel for handling the online customer service problem; According to the handling ability score of the customer service personnel and the number of customer service problems currently handled by the customer service personnel, determining a comprehensive score of the customer service personnel; Based on the comprehensive score and the number of customer service problems, determining the customer service personnel for handling the online customer service problem.

2. The method according to claim 1, wherein The obtaining a handling ability score of a customer service personnel for handling the online customer service problem based on the first keyword and a personnel ability scoring model includes: In the personnel ability scoring model, determining a second keyword that matches the first keyword; According to the corresponding relationship between the second keyword and the handling ability score of the customer service personnel, obtaining the handling ability score.

3. The method according to claim 2, wherein Constructing the personnel ability scoring model includes: Obtaining an information interaction document of the customer service problems handled by the customer service personnel, and determining a second keyword corresponding to the information interaction document; Determining a handling ability score of the customer service personnel for handling the customer service problem corresponding to the second keyword; According to the corresponding relationship between the second keyword and the handling ability scores of each customer service personnel, constructing the personnel ability scoring model.

4. The method according to claim 3, wherein, The determining a handling ability score of the customer service personnel for handling the customer service problem corresponding to the second keyword includes: Determining a weight score of the second keyword; Obtaining the satisfaction and duration of the customer service personnel for handling the customer service problem corresponding to the second keyword; According to the weight score, the satisfaction, and the duration, determining the handling ability score of the customer service personnel for handling this customer service problem.

5. The method according to claim 4, wherein, The determining a weight score of the second keyword includes: Determining the number of occurrences of the second keyword in the information interaction document; According to the number of occurrences and the length information of the information interaction document, determining the occurrence frequency of the second keyword; Based on the number of information interaction documents containing the second keyword and the total number of information interaction documents, determining the inverse document frequency of the second keyword; According to the occurrence frequency, the inverse document frequency, the length of the information interaction document, the average length of all information interaction documents, and an adjustment factor, determining the weight score of the second keyword.

6. The method according to any one of claims 3 to 5, wherein, The determining a second keyword corresponding to the information interaction document includes: Performing word segmentation processing on the information interaction document to obtain a plurality of first word segments; Determining the number of occurrences and weight information of each first word segment; According to the number of occurrences, the weight information of each first word segment, a damping coefficient, and the total number of the first word segments, determining the keyword score of each first word segment; Based on the keyword scores of each first word segment, determining the second keyword.

7. The method according to claim 6, wherein, The based on the keyword scores of each first word segment, determining the second keyword includes: Among the keyword scores of each first participle, at least one second participle with the highest keyword score is selected as the second keyword.

8. The method according to claim 1, wherein, The determining of the comprehensive score of the customer service staff according to the processing ability score of the customer service staff and the number of customer service problems currently processed by the customer service staff includes: Obtain the processing ability score of the customer service staff as the processing ability weight value; Based on the processing ability weight value and the number of customer service problems, calculate the proportion of the processing ability weight value of the customer service staff as the comprehensive score; The determining of the customer service staff for processing the online customer service problem based on the comprehensive score and the number of customer service problems includes: According to the proportion of the processing ability weight values of all customer service staff, select the customer service staff for processing the online customer service problem among all customer service staff.

9. The method according to claim 8, wherein, The selecting of the customer service staff for processing the online customer service problem according to the proportion of the processing ability weight values of all customer service staff includes: Among the proportions of the processing ability weight values of all customer service staff, select the minimum proportion of the processing ability weight value; Based on the minimum proportion of the processing ability weight value, select the customer service staff for processing the online customer service problem.

10. The method according to claim 9, wherein, The selecting of the customer service staff for processing the online customer service problem based on the minimum proportion of the processing ability weight value includes: If the number of the minimum proportion of the processing ability weight value is 1, then use the customer service staff corresponding to this minimum proportion of the processing ability weight value as the customer service staff for processing the online customer service problem; If the number of the minimum proportion of the processing ability weight value is multiple, then select the customer service staff with the least number of currently processed customer service problems among the multiple customer service staff corresponding to the multiple minimum proportions of the processing ability weight value as the customer service staff for processing the online customer service problem.

11. The method according to claim 1, wherein The extracting of the first keyword from the problem information includes: Perform word segmentation processing on the problem information to obtain multiple second participles; Determine the occurrence times and weight information of each second participle; According to the occurrence times, weight information of each second participle, as well as the damping coefficient and the total number of the second participles, determine the keyword score of each second participle; Based on the keyword scores of each second participle, determine the first keyword.

12. The method according to claim 11, wherein, The determining of the first keyword based on the keyword scores of each second participle includes: Among the keyword scores of each second participle, select the second participle with the highest keyword score as the first keyword.

13. The method according to claim 1, further comprising: After the online problem is processed, perform an update process on the personnel ability scoring model.

14. A customer service problem processing device, comprising: An information extraction module, configured to obtain the problem information of the online customer service problem and extract the first keyword from the problem information when receiving a request for manually processing the online customer service problem; An ability scoring module, configured to obtain the processing ability score of the customer service staff for processing the online customer service problem based on the first keyword and the personnel ability scoring model; A comprehensive scoring module, configured to determine a comprehensive score of the customer service staff according to the processing ability score of the customer service staff and the number of customer service problems currently processed by the customer service staff; A personnel allocation module, configured to determine the customer service staff for processing the online customer service problems based on the comprehensive score and the number of customer service problems.

15. A customer service problem processing device, comprising: A memory; And a processor coupled to the memory, the processor being configured to execute the method according to any one of claims 1 to 13 based on instructions stored in the memory.

16. A computer-readable storage medium storing computer instructions, the instructions being executed by a processor to perform the method according to any one of claims 1 to 13.

Images