CN106228286A - A kind of data analysing method for the assessment of artificial customer service work quality - Google Patents

Abstract

The invention discloses a kind of data analysing method for the assessment of artificial customer service work quality, the method is based on fuzzy synthetic appraisement method, evaluate the practical situation of system according to customer service, from representativeness, integrity and the suitability angularly, set up the comprehensive fuzzy evaluation index system of artificial customer service；Use principal component analytical method to determine the weight of different index, thus form the weight set of evaluation index；5 grades of scales are used to set up opinion rating class collection；Carry out single factor test fuzzy evaluation, determine appraisal fuzzy matrix；Use operator that synthesis computing is carried out fuzzy overall evaluation, obtain the fuzzy subset on opinion rating class collection；Finally carry out the normalized of comprehensive fuzzy evaluation index, determine final evaluation of estimate.Compared with prior art, the present invention proposes the fuzzy evaluation model improved based on principal component analysis, objective by data analysing method, be efficiently completed customer service performance evaluation, thus help customer service management personnel preferably to find problem and shortage present in work from data, realize the lifting of service quality and CSAT with a definite target in view.

Description

Data analysis method for artificial customer service work quality assessment

Technical Field

The application relates to the field of data processing, in particular to a data analysis method aiming at the evaluation of the working quality of artificial customer service.

Background

With the increasingly intense market competition of enterprises, the way of providing services for customers by each large enterprise gradually develops from the traditional single channel to the diversified channel, such as 95598 customer service centers of national power grids, electronic channels of communication operators, online banks of banks and the like. In many service channels, manual customer service is a way to realize one-to-one effective service, and especially when completing the work of customer service such as customer service for core high-end customers, sales-type or shopping-guide type customer service, complaint and the like which need emotional communication, and complex business explanation type customer service, the manual customer service is a very important and effective method, so the manual customer service becomes a necessary way for each enterprise to establish a customer service system at present.

With the continuous development and growth of enterprises, the number of seats served by human customers generally increases exponentially to meet the increasing customer demands. The scale of the customer service center is continuously enlarged, along with the difficulty of the management work of the customer service quality, managers are difficult to find ways for effectively improving the work quality in a large amount of customer service work.

In the traditional work evaluation method for customer service personnel, professional scoring personnel are mostly adopted to finish manual scoring. The evaluation method of the mode has more human factors, and particularly under the background that the scale of a customer service center is larger and larger, objective problems which are not noticed by individuals but exist are easily ignored, and the method cannot well adapt to the requirements of large-scale manual customer service work quality management and improvement at present.

Disclosure of Invention

In view of the above, the present application provides a data analysis method for manual customer service work quality evaluation, which objectively and efficiently completes customer service work evaluation through the data analysis method, thereby helping customer service managers to better find problems and deficiencies existing in work from data, and purposefully realizing improvement of service quality and customer satisfaction.

In order to achieve the above object, the following is proposed:

a data analysis method for manual customer service work quality assessment comprises the following steps:

establishing an index system of fuzzy comprehensive evaluation, taking U as a set consisting of various main factors influencing an evaluation object (artificial customer service working quality): u ═ U₁，u₂，u₃，……，u_n}；

Determining the weight of the index by adopting a principal component analysis method, and forming a weight set A ═ a of the evaluation index₁，a₂，a₃，……，a_nAssigning corresponding weight values a to the factors according to the importance degrees of the factors_i；

Establishing an evaluation grade set V by adopting a multi-grade scale, wherein V represents the grade of the evaluation factors, and V is { V ═ V }₁，v₂，v₃，……，v_m}；

Performing single-factor fuzzy evaluation according to a single index in the indexes to determine an evaluation fuzzy matrix, wherein the single factor u_i(i 1, 2, … …, n) to obtain a fuzzy subset on V, i.e. a one-factor evaluation vector R_i＝(r_i1，r_i2，r_i3，……，r_im) Thus determining a fuzzy evaluation matrix from U to V, r_ijIndicating index u_iEvaluated asv_jOf the degree of membership of, wherein,

$R={\left(r_{ij}\right)}_{n\×m}=\left[\begin{array}{cccc}{r}_{11}& r_{12}& \mathrm{...}& r_{1m}\\ r_{21}& r_{22}& \mathrm{...}& r_{2m}\\ \mathrm{...}& \mathrm{...}& \mathrm{...}& \mathrm{...}\\ r_{n1}& r_{n2}& \mathrm{...}& r_{nm}\end{array}\right]$

weight value a for each evaluation index_iRelative membership r to each respective evaluation index_nmPerforming an operator pair synthesis operation to obtain a fuzzy evaluation comprehensive index B ═ A > R ═ B₁，b₂，b₃，……，b_m) I.e. by

$B=(a_1,a_2,a_3,...,a_n)\left[\begin{array}{cccc}{r}_{11}& r_{12}& \mathrm{...}& r_{1m}\\ r_{21}& r_{22}& \mathrm{...}& r_{2m}\\ \mathrm{...}& \mathrm{...}& \mathrm{...}& \mathrm{...}\\ r_{n1}& r_{n2}& \mathrm{...}& r_{nm}\end{array}\right]=(b_1,b_2,b_3,...,b_m)$

Finally, the normalization processing of the fuzzy comprehensive evaluation index is carried out to determine the final evaluation value,

in a preferred embodiment of the present invention, the synthesis operation of a and R may employ an operator pair: m (A, V-shaped),

wherein, the 'A' represents that the two are compared and then the small value is obtained, and the 'V' represents that the two are compared and then the large value is obtained.

In a preferred embodiment of the present invention, the evaluation result indicator is processed by a weighted average method. With b_jAre weighted values and are normalized, i.e.Element V for each evaluation set_jA weighted average is performed, the result being the final evaluation, i.e.The larger the comprehensive index value vj is, the more excellent the jth object is, and scientific judgment can be performed according to the situation.

In a preferred embodiment of the present invention, the main factors include 4 dimensions of workload, work efficiency, customer satisfaction, attendance.

In a preferred embodiment of the present invention, the weight value a_iSatisfies the following conditions:and a is_i>0。

In a preferred embodiment of the present invention, the grade rank set V, the larger the score is, the higher the representative evaluation is, a 5-grade scale is used to establish the evaluation grade rank set, that is, V ═ excellent, good, general, poor, and very poor } is converted into a corresponding numerical type of V ═ 5, 4, 3, 2, 1}, and then the original evaluation data set of each collected artificial customer service is obtained

The covariance matrix of the original evaluation dataset is calculated again, ∑ ═ σ_ij)_P×PWherein,

${\mathrm{\σ}}_{ij}=\frac{1}{n-1}\underset{k=1}{\overset{n}{\Σ}}(x_{ki}-\stackrel{\‾}{x_i})(x_{kj}-\stackrel{\‾}{x_j}),i,j=1,2,\mathrm{...},p$

determining eigenvalues λ of covariance ∑₁≥λ₂≥……≥λ_p>0 and corresponding orthogonalized unit feature vector U_i＝(u_1iu_2i…u_pi)^T

The ith principal component of X is F_i＝U_iX，i＝1,2,……，p。

Using variance contribution ratioExplain principal component F_iThe size of the amount of information reflected,is F₁,F₂,……,F_l(l<P) of the variance contribution rate.

Reasonably selecting l from all the determined p main components to realize the final evaluation analysis. And extracting principal components with characteristic roots >1 corresponding to the principal components and cumulative variance contribution rate of the first N principal components more than 80%, and taking the principal components as the finally determined one principal component.

And dividing the load number of the determined principal component to the original index by the square root of the characteristic root corresponding to the principal component to determine the coefficient of each index in the linear combination of the N principal components. And carrying out weighted average on the newly obtained coefficient by utilizing the determined variance contribution rate of each principal component, and then carrying out normalization processing on the obtained system to obtain a final weight set A of each evaluation index.

Further, the fuzzy comprehensive evaluation is used for establishing a model of a connection hierarchical structure for each element of the complex system to be evaluated according to the association membership of the element, constructing a judgment matrix, and solving the weight of each element and the consistency of the check and correction judgment matrix according to the judgment matrix.

By the technical scheme, the data analysis method applied to the manual customer service work evaluation is disclosed. The method is based on a fuzzy comprehensive evaluation method, and a fuzzy comprehensive evaluation index system of artificial customer service is established from the aspects of representativeness, completeness, applicability and the like according to the actual situation of a customer service evaluation system; determining the weights of different indexes by adopting a principal component analysis method so as to form a weight set of evaluation indexes; establishing an evaluation grade level set by adopting a 5-grade scale; performing single-factor fuzzy evaluation to determine an evaluation fuzzy matrix; carrying out fuzzy comprehensive evaluation on the synthetic operation by adopting an operator to obtain a fuzzy subset on the evaluation grade level set; and finally, carrying out normalization processing on the fuzzy comprehensive evaluation index to determine a final evaluation value. Compared with the prior art, the invention provides the fuzzy evaluation model based on principal component analysis improvement, and completes customer service work evaluation objectively and efficiently through a data analysis method, thereby helping customer service managers to better find problems and defects existing in work from data, and purposefully realizing the improvement of service quality and customer satisfaction.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below. It is obvious that the drawings in the following description are only embodiments of the invention, and that for a person skilled in the art, other drawings can be obtained from the provided drawings without inventive effort.

FIG. 1 is a flow chart diagram illustrating a data processing method applied to a human customer service work evaluation system according to an embodiment of the present invention;

FIG. 2 is a flow chart illustrating a data processing method applied to a human customer service work evaluation system according to another embodiment of the present disclosure;

FIG. 3 is a flow chart illustrating a data processing method applied to a human customer service work evaluation system according to another embodiment of the present disclosure;

FIG. 4 is a flow chart illustrating a data processing method applied to a human customer service work evaluation system according to another embodiment of the present disclosure;

FIG. 5 is a flow chart illustrating a data processing method applied to a human customer service work evaluation system according to another embodiment of the present disclosure.

Detailed Description

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.

Referring to fig. 1, a flow chart of a data processing method applied to a human customer service work evaluation system according to an embodiment of the present invention is shown.

The method comprises the following steps:

s101: establishing an index system of fuzzy comprehensive evaluation, taking U as a set consisting of various main factors influencing an evaluation object (artificial customer service working quality): u ═ U₁，u₂，u₃，……，u_n}

When evaluating the manual customer service staff, the manual customer service staff are generally graded from multiple dimensions, and each angle is provided with multiple preset evaluation indexes. The method is mainly used for evaluating 4 dimensions of workload, working efficiency, customer satisfaction and attendance check. Wherein the evaluation indicators contained in the workload dimension are: the contribution degree of the workload to the team and the ranking of the workload team. The evaluation indicators included in the work efficiency dimension are: the contribution degree of the work efficiency to the team and the team cooperation degree. The evaluation indicators included in the customer satisfaction dimension are: service evaluation satisfaction, service complaint poor evaluation and team member mutual evaluation excellent show. The evaluation indexes included in the attendance check dimension are: attendance ranking and order making ranking.

S102: according to the importance degree of each factor, corresponding weight value a is given to each factor_iThus, a weight set a of evaluation factors is composed of { a ═ a }₁，a₂，a₃，……，a_nAnd (4) judging consistency check and correction of the matrix A and calculation of the weight of the matrix A, wherein the requirements are as follows:and a is_i>0. The determination of the weight set may be determined using a principal component analysis method.

S103: an evaluation grade rank set V is established. V represents the rank level of the team's evaluation factor, V ═ V₁，v₂，v₃，……，v_mIn the method, the grades are classified into 5 grades, and V is { excellent, good, general, poor, and bad }.

S104: and performing single-factor fuzzy evaluation to determine an evaluation fuzzy matrix. For single factor u_i(i 1, 2, … …, n) to obtain a fuzzy subset on V, i.e. a one-factor evaluation vector R_i＝(r_i1，r_i2，r_i3，……，r_im) Thus, a fuzzy evaluation matrix from U to V is determined. r is_ijIndicating index u_iRated as v_jDegree of membership. Wherein,

$R={\left(r_{ij}\right)}_{n\&times;m}=\left[\begin{array}{cccc}{r}_{11}& r_{12}& \mathrm{...}& r_{1m}\\ r_{21}& r_{22}& \mathrm{...}& r_{2m}\\ \mathrm{...}& \mathrm{...}& \mathrm{...}& \mathrm{...}\\ r_{n1}& r_{n2}& \mathrm{...}& r_{nm}\end{array}\right]$

s105: weight value a for each evaluation index_iRelative membership r to each respective evaluation index_nmAnd carrying out operator pair synthesis operation to obtain a comprehensive index B of fuzzy evaluation.

Considering the weight value distribution under the condition of multiple factors, the fuzzy comprehensive evaluation model of the artificial customer service working quality is as follows: b ═ ao R ═ B ═ R ═ B₁，b₂，b₃，……，b_m) Namely:

$B=(a_1,a_2,a_3,...,a_n)\left[\begin{array}{cccc}{r}_{11}& r_{12}& \mathrm{...}& r_{1m}\\ r_{21}& r_{22}& \mathrm{...}& r_{2m}\\ \mathrm{...}& \mathrm{...}& \mathrm{...}& \mathrm{...}\\ r_{n1}& r_{n2}& \mathrm{...}& r_{nm}\end{array}\right]=(b_1,b_2,b_3,...,b_m)$

and carrying out U-to-V fuzzy transformation ao R, namely synthesis operation, to obtain a fuzzy subset B on V, namely a fuzzy comprehensive evaluation result of the evaluation object.

The synthesis operation of A and R can adopt an operator pair: m (A, V-V).

Wherein, the 'A' represents that the two are compared and then the small value is obtained, and the 'V' represents that the two are compared and then the large value is obtained.

S106: and processing the evaluation result index by adopting a weighted average method. With b_jAre weighted values and are normalized, i.e.For each evaluation set element v_jA weighted average is performed, the result being the final evaluation, i.e.

From the above examples it can be seen that: the application discloses a data analysis method applied to manual customer service work evaluation. The method is based on a fuzzy comprehensive evaluation method, and a fuzzy comprehensive evaluation index system of artificial customer service is established from the aspects of representativeness, completeness, applicability and the like according to the actual situation of a customer service evaluation system; determining the weights of different indexes by adopting a principal component analysis method so as to form a weight set of evaluation indexes; establishing an evaluation grade level set by adopting a 5-grade scale; performing single-factor fuzzy evaluation to determine an evaluation fuzzy matrix; carrying out fuzzy comprehensive evaluation on the synthetic operation by adopting an operator to obtain a fuzzy subset on the evaluation grade level set; and finally, carrying out normalization processing on the fuzzy comprehensive evaluation index to determine a final evaluation value. Compared with the prior art, the invention provides the fuzzy evaluation model based on principal component analysis improvement, and completes customer service work evaluation objectively and efficiently through a data analysis method, thereby helping customer service managers to better find problems and defects existing in work from data, and purposefully realizing the improvement of service quality and customer satisfaction.

Referring to fig. 2, a flow chart of a data processing method applied to a human customer service work evaluation system according to another embodiment of the present invention is shown.

The method specifically comprises the following steps:

s201: for the evaluation grade level set V, the larger the score is, the higher the evaluation is, that is, V is { excellent, good, generally poor, very bad }, the conversion is to a corresponding numerical type of V {5, 4, 3, 2, 1}, and then the original evaluation data set of each collected artificial customer service is obtained

Specifically, the method comprises the following steps of: and the evaluation grade grades corresponding to the contribution degree of the workload to the teams, the workload team ranking, the contribution degree of the work efficiency to the teams, the team cooperation degree, the service evaluation satisfaction degree, the service complaint poor evaluation degree, the team member mutual evaluation goodness and elegance degree, the attendance ranking and the order making ranking are converted into corresponding numerical types. The specific conversion results are given in the following table:

s202, calculating the covariance matrix of the original evaluation data set, ∑ ═ σ_ij)_P×PWherein,

${\mathrm{\&sigma;}}_{ij}=\frac{1}{n-1}\underset{k=1}{\overset{n}{\&Sigma;}}(x_{ki}-\stackrel{\&OverBar;}{x_i})(x_{kj}-\stackrel{\&OverBar;}{x_j}),i,j=1,2,\mathrm{...},p$

determining eigenvalues λ of covariance ∑₁≥λ₂≥……≥λ_p>0 and corresponding orthogonalized unit feature vector U_i＝(u_1iu_2i…u_pi)^T

The ith principal component of X is F_i＝U_iX，i＝1,2,……，p。

Using variance contribution ratioExplain principal component F_iThe size of the amount of information reflected,is F₁,F₂,……,F_l(l<P) of the variance contribution rate.

Reasonably selecting l from all the determined p main components to realize the final evaluation analysis. And extracting principal components with characteristic roots >1 corresponding to the principal components and the cumulative variance contribution rate of the first N principal components being more than 80%, and taking the principal components as finally determined principal components.

S203: the load number of the determined principal component to the original index, namely the corresponding orthogonalization unit characteristic vector U, is divided by the square root of the characteristic root lambda corresponding to the principal component to determine the coefficient of each index in the linear combination of the N principal components. And performing weighted average on coefficients in the principal component linear combination by using the determined variance contribution rate of each principal component, and performing normalization processing on the obtained system to obtain a final weight set A of each evaluation index.

Specifically, in the method, the index weight is equal to the variance contribution rate of the principal component as the weight, and the index weight is calculated based on the weightThe result of normalization of the weighted average of the coefficients in each principal component linear combination is indicated. Therefore, first, coefficients of the indexes in each principal component linear combination need to be determined, and a specific calculation method is as follows:the newly obtained linear combination of principal components is F_i＝χ₁X-χ₂X+……+χ_nX。

Then, based on the variance contribution rate of the principal component, weighted average is carried out on the coefficients in the newly obtained linear combination of the principal component, namelyThe resultant linear score equation obtained at this time is Y ═ μ₁X-μ₂X+……+μ_nX。

And finally, normalizing the indexes of the equation. The normalization method comprises the following steps:the final weight equation of each evaluation index is Y- ξ₁X-ξ₂X+……+ξ_nX。

In this example, the equation of each finally determined evaluation index weight is: evaluation score of manual customer service work of 13.36%. work contribution to team + 10.18%. work contribution + 10.87%. work efficiency + 7.42%. team cooperation + 14.93%. service evaluation satisfaction + 8.18%. service complaint poor evaluation + 12.33%. team member mutual evaluation excellent evaluation + 9.27%. attendance + 13.46%. system of single ranking

Referring to fig. 3, a flow chart of a data processing method applied to a human customer service work evaluation system according to another embodiment of the present invention is shown.

The method specifically comprises the following steps:

s301: and counting the quantity of each evaluation index evaluated as different evaluation grade grades to form a statistical matrix.

S302: and calculating the membership degree of each evaluation index in different evaluation grade grades based on the statistical matrix. The calculation method is as follows:

s303: and for each customer service person, establishing a single-factor fuzzy evaluation matrix thereof, and finally determining a fuzzy evaluation matrix R from U to V.

Referring to fig. 4, a flow chart of a data processing method applied to a human customer service work evaluation system according to another embodiment of the present invention is shown.

The method specifically comprises the following steps:

considering the weight value distribution under the condition of multiple factors, the fuzzy comprehensive evaluation model of the artificial customer service working quality is as follows: b ═ ao R ═ B ═ R ═ B₁，b₂，b₃，……，b_m) Namely:

$B=(a_1,a_2,a_3,...,a_n)\left[\begin{array}{cccc}{r}_{11}& r_{12}& \mathrm{...}& r_{1m}\\ r_{21}& r_{22}& \mathrm{...}& r_{2m}\\ \mathrm{...}& \mathrm{...}& \mathrm{...}& \mathrm{...}\\ r_{n1}& r_{n2}& \mathrm{...}& r_{nm}\end{array}\right]=(b_1,b_2,b_3,...,b_m)$

and carrying out U-to-V fuzzy transformation ao R, namely synthesis operation, to obtain a fuzzy subset B on V, namely a fuzzy comprehensive evaluation result of the evaluation object.

The synthesis operation of A and R can adopt an operator pair: m (A, V-V).

Wherein, the 'A' represents that the two are compared and then the small value is obtained, and the 'V' represents that the two are compared and then the large value is obtained.

The fuzzy transformation process of U to V is carried out on the synthesis operation through an operator as follows:

s401: and taking a small calculation. Each number in a is compared with the 1 st, 2 nd, … … th and nth rows of matrix R in turn, and the smaller of the two values is retained, forming a new matrix b.

S402: and taking a big calculation. In the matrix B, the maximum value of each column is selected in turn to form the fuzzy subset B on the final V.

Referring to fig. 5, a flow chart of a data processing method applied to a human customer service work evaluation system according to another embodiment of the present invention is shown.

The method specifically comprises the following steps:

s501: with b_jFor weighting values, normalization is performed, i.e.

If the final fuzzy comprehensive evaluation result of the artificial customer service V001 on V is (0.35,0.35,0.1374,0.024,0) in this example, the evaluation result isThe fuzzy comprehensive evaluation result B after the normalization processing is (0.4063,0.4063,0.1595,0.0279, 0).

S502: the evaluation grade set V is numerically specified as excellent, good, general, poor, very poor, resulting in V being {1.00, 0.85, 0.70, 0.50, 0.00 }. Elements v of each evaluation set based on fuzzy comprehensive evaluation result B_jPerforming a weighted average, i.e.The results obtained are the final evaluations.

As in this example, the fuzzy comprehensive evaluation result of the customer service manual V001 (0.4063 × 1.00+0.4063 × 0.85+0.1595 × 0.70+0.0279 × 0.50+0 × 0.00) — 0.87727 was at a good level.

Finally, it should also be noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.

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

Claims (6)

1. A data analysis method aiming at the evaluation of the working quality of artificial customer service is characterized by comprising the following steps:

establishing an index system of fuzzy comprehensive evaluation, taking U as a set consisting of various main factors influencing an evaluation object (artificial customer service working quality): u ═ U₁，u₂，u₃，……，u_n}；

Determining the weight of the index by adopting a principal component analysis method, and forming a weight set A ═ a of the evaluation index₁，a₂，a₃，……，a_nAssigning corresponding weight values a to the factors according to the importance degrees of the factors_i；

Establishing an evaluation grade set V by adopting a multi-grade scale, wherein V represents the grade of the evaluation factors, and V is { V ═ V }₁，v₂，v₃，……，v_m}；

Performing single-factor fuzzy evaluation according to a single index in the indexes to determine an evaluation fuzzy matrix, wherein the single factor u_i(i 1, 2, … …, n) to obtain a fuzzy subset on V, i.e. a one-factor evaluation vector R_i＝(r_i1，r_i2，r_i3，……，r_im) Thus determining a fuzzy evaluation matrix from U to V, r_ijIndicating index u_iRated as v_jOf the degree of membership of, wherein,

weight value a for each evaluation index_iRelative membership r to each respective evaluation index_nmPerforming an operator pair synthesis operation to obtain a fuzzy evaluation comprehensive index B ═ A > R ═ B₁，b₂，b₃，……，b_m) I.e. by

And finally, carrying out normalization processing on the fuzzy comprehensive evaluation index to determine a final evaluation value.

2. The data analysis method for manual customer service work quality assessment according to claim 1,

the synthesis operation of A and R can adopt an operator pair: m (A, V-shaped),

wherein, the 'A' represents that the two are compared and then the small value is obtained, and the 'V' represents that the two are compared and then the large value is obtained.

3. The data analysis method for the evaluation of the quality of the manual customer service work according to claim 2, wherein the evaluation result index is processed by a weighted average method. With b_jAre weighted values and are normalized, i.e.Element V for each evaluation set_jA weighted average is performed, the result being the final evaluation, i.e.The larger the comprehensive index value vj is, the more excellent the jth object is, and scientific judgment can be performed according to the situation.

4. The data analysis method for manual customer service work quality assessment according to claim 1, wherein the main factors include 4 dimensions of workload, work efficiency, customer satisfaction and attendance assessment.

5. The method as claimed in claim 1, wherein the weight value a is a_iSatisfies the following conditions:and a is_i>0。

6. The data analysis method for the evaluation of the quality of the customer service work according to claim 1, wherein the grade set V represents higher evaluation with higher score, and a 5-grade scale is used to establish the evaluation grade set, i.e. V ═ { excellent, good, general, poor } is converted into V ═ 5, 4, 3, 2, 1} corresponding to the numerical type.