KR101824242B1 - Apparatus for estimating total counsel quality - Google Patents

Abstract

The present invention discloses a counsel quality total evaluation device which allows a counselor to compare counsel data calling with a customer with a pre-stored evaluation reference model and evaluates counsel quality of the entire counsel data. The counsel quality total evaluation device of the present invention comprises: an evaluation data collection module collecting evaluation data to be subjected to counsel quality evaluation; a counsel quality evaluation module analyzing the evaluation data collected by the evaluation data collection module, and evaluating the counsel quality for each predetermined evaluation item; and an output part outputting a counsel quality evaluation result evaluated by the counsel quality evaluation module.

Description

[0001] APPARATUS FOR ESTIMATING TOTAL COUNCEL QUALITY [0002]

More particularly, the present invention relates to a consultation quality evaluation apparatus for evaluating the consultation quality of all consultation data by comparing the consultation data that the consultant has made with the customer to the previously stored evaluation reference model, .

In recent years, customer relationship management (CRM) has become the center of marketing in order to meet the needs of diversified customers. In many companies, call centers are operated regardless of the type and size of business.

The call center serves to provide various information while the counselor responds to the customer by telephone, and receives and solves inquiries and inconveniences of customers. In addition, consultation and trouble-shooting via the Internet are also being conducted due to the development of the Internet. However, due to the speed of the processing and the ease of access, consultation using the call center is still taking a high proportion.

In order to provide better quality counseling service to the call center, the counseling counselor should be monitored periodically and the counselor should be evaluated.

However, in the conventional call center system, the contents of consultation of counselors are simply recorded and information about each consultation is simply stored. In order to monitor or evaluate the consultation contents of the counselor client, the evaluator has to extract and confirm the data individually.

That is, conventionally, there is a problem in that the voice file and data stored in the consultation contents recording server or the consultation history database can not be processed and used for evaluation, so that the counseling client can not be effectively monitored or evaluated.

Background Art [0002] The background art of the present invention is disclosed in Korean Patent Registration No. 10-0507325 (Aug. 1, 2005) entitled " Method and system for evaluating agent education through a network ".

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems, and an object of one aspect of the present invention is to provide a method and system for evaluating the quality of counseling of entire consultation data by comparing counseling data, And to provide a consultation quality inheritance evaluating apparatus.

According to one aspect of the present invention, there is provided an apparatus for evaluating the quality of consultation quality, comprising: an evaluation data collection module for collecting assessment data to be consultation quality evaluation; A consultation quality evaluation module for analyzing the evaluation data collected by the evaluation data collection module and evaluating the consultation quality for each predetermined evaluation item; And an output unit for outputting a consultation quality evaluation result evaluated by the consultation quality evaluation module.

The evaluation data collection module of the present invention includes: a first evaluation data collection unit for collecting the post-processing classification data created by the consultant after consultation between the customer and the consultant; And a second evaluation data collection unit for collecting STT (Speech To Text) data obtained by converting a consultation call between the customer and the consultant into text.

The STT data of the present invention includes at least one of talk time, silence time, consultation text, and voice waveform characteristics.

The consultation quality evaluation module of the present invention includes: a classification unit for classifying all counseling calls into a predetermined business domain; And a consultation quality evaluating unit for evaluating the consultation quality of the evaluation data according to the evaluation items according to the evaluation reference model after detecting a predetermined evaluation reference model according to the business area classified in the classification unit .

The business domain of the present invention includes at least one of a product operated by an organization and a service provided to a customer by an organization.

The consultation quality evaluation module of the present invention compares the evaluation data with the business knowledge evaluation model of the evaluation reference model corresponding to the business area and evaluates the business knowledge level of the consultant according to the comparison result. A talk time analyzer for comparing the evaluation data with the talk time evaluation model of the evaluation reference model corresponding to the business area and analyzing the talk time of the consultant according to the comparison result; An o consultation reason analyzer for comparing the evaluation data with a misclassification analysis model of the evaluation criterion model corresponding to the work area and analyzing a consultant's reason for misconduct according to the comparison result; A customer propensity classifying unit for comparing the evaluation data with a sensitivity analysis model of the evaluation reference model corresponding to the business area and classifying the customer propensity according to the comparison result; And a consultation type classification unit for comparing the post-treatment classification history model of the evaluation reference model with the consultation type and the post-treatment classification details directly created by the consultant after the consultation call and classifying the consultation type according to the comparison result And a type classification unit.

The business knowledge level evaluating unit of the present invention analyzes the similarity between the evaluation data and the business knowledge evaluation model of the evaluation reference model corresponding to the business area and determines the business knowledge level of the consultant based on the emotion analysis result of the evaluation data .

The business knowledge evaluation model of the evaluation criterion model of the present invention is characterized by including at least one of a predefined consultation script, a frequently asked questions (FAQ), and misexcrement information.

The business knowledge level evaluating unit of the present invention is characterized by warning that the evaluation of the business knowledge level can not be performed on the evaluation data when the noise level of the evaluation data is equal to or higher than the setting level or when the evaluation data is out of the business area.

The talk time analysis unit of the present invention extracts the silence time from the evaluation data through the time interval between the customer's question and the response time of the consultant, and if the extracted silence time is more than the set silence time of the talk time evaluation model, And analyzing reasons of silence of the counselor through whether or not it matches the predetermined setting pattern.

The talk time analysis unit of the present invention is characterized in that the talk time analysis can not be performed on the evaluation data when the noise level of the evaluation data is equal to or higher than the set level or when the evaluation data is out of the business area.

The o consultation reason analyzing unit of the present invention is characterized by comparing the evaluation data with the similarity of the o consultation analysis model corresponding to the job domain and analyzing the reasons of the o consultation of the consultant according to the comparison result.

The o consultation reason analyzing unit of the present invention is characterized in that if the noise level of the evaluation data is equal to or higher than a set level or if the evaluation data is out of the business area,

The customer propensity classifier of the present invention classifies the propensity of the customer into one of positive tendency and negative tendency based on the emotional analysis result of the evaluation data.

The customer propensity classifier of the present invention is characterized in that the customer propensity classifying section warns the customer that propensity classification can not be performed on the evaluation data if the noise level of the evaluation data is equal to or higher than a set level or the evaluation data deviates from the business area.

The consultation type classification unit of the present invention compares the post-processing classification history model of the evaluation criterion model of the consultation type with the post-processing classification history of the consultant after the consultation call and compares the post- Or not.

The consultation type classifying unit of the present invention alerts the user that the consultation type can not be classified for the evaluation data if the noise level of the evaluation data is equal to or higher than a set level or the evaluation data is out of the business area.

The output unit of the present invention performs statistical processing on the output of the business knowledge level evaluating unit, the talk time analyzing unit, the ocontact reason analyzing unit, the customer propensity classifying unit, and the consultation type classifying unit by counselor, And a statistical processing unit.

According to one aspect of the present invention, the consultation quality evaluation apparatus evaluates the consultation quality of the entire consultation data by comparing the consultation data that the consultant has made with the customer to the stored evaluation reference model.

According to another aspect of the present invention, the consultation quality transfer evaluation device can be utilized for training planning and education for a counselor by evaluating the consultation quality of counseling materials in various aspects such as business knowledge, talk time, false consultation, and emotional evaluation .

The consultation quality transfer evaluation apparatus according to another aspect of the present invention classifies the customer orientation based on the sensitivity evaluation, thereby enabling the consultation service for the customer to be performed more efficiently.

According to still another aspect of the present invention, the consultation quality transfer evaluation device provides the consultation quality evaluation result and the business status in real time, so that the management of the consultation quality can be performed in real time.

1 is a block diagram of a consultation quality inherent evaluation apparatus according to an embodiment of the present invention.
2 is a flowchart of a consultation quality inheritance evaluation method according to an embodiment of the present invention.
FIG. 3 is a view showing a real-time monitoring status of consultation quality evaluation results according to an embodiment of the present invention.
FIG. 4 is a view showing a consultation quality analysis detailed screen according to an embodiment of the present invention.
FIG. 5 is a screen showing a result of consultation quality analysis by a consultant in accordance with an embodiment of the present invention.
6 is a view showing an evaluation management screen according to an embodiment of the present invention.
FIG. 7 is a diagram illustrating a screen for a query condition and an evaluation history list according to an exemplary embodiment of the present invention.
FIG. 8 is a diagram illustrating a screen for evaluation statistics according to a query condition according to an embodiment of the present invention.
FIG. 9 is a view showing a screen for the consultation type transition and the consulting general index according to an embodiment of the present invention.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, a consultation quality evaluation apparatus according to an embodiment of the present invention will be described in detail with reference to the accompanying drawings. In this process, the thicknesses of the lines and the sizes of the components shown in the drawings may be exaggerated for clarity and convenience of explanation. Further, the terms described below are defined in consideration of the functions of the present invention, which may vary depending on the user, the intention or custom of the operator. Therefore, definitions of these terms should be made based on the contents throughout this specification.

1 is a block diagram of a consultation quality inherent evaluation apparatus according to an embodiment of the present invention.

1, the consultation quality inherent evaluation apparatus according to an embodiment of the present invention includes an evaluation data collection module 10, a consultation quality evaluation module 20, an output unit 30, and an administrator terminal 40 .

The evaluation data collection module 10 collects evaluation data to be subjected to consultation quality evaluation.

The evaluation data includes STT data converted into Speech To Text (STT) data after consultation between the client and the counselor, and the consultation call between the client and the counselor. In addition, the evaluation data may include any data that can be used to evaluate the quality of the consultation.

The evaluation data collection module 10 includes a first evaluation data collection unit 11 and a second evaluation data collection unit 12.

The first evaluation data collection unit 11 collects the post-processing classification data created by the consultant. When the consultation call with the customer is completed, the counselor creates a post-processing classification breakdown according to the consultation call. Then, the first evaluation data collection unit 11 collects the post-processing classification details from the counseling terminal (not shown) of each consultant and inputs them to the consultation quality evaluation module 20. [

The second evaluation data collection unit 12 converts the consultation call between the client and the consultant into STT data to generate STT data, and inputs the generated STT data to the consultation quality evaluation module 20. [

The STT data may include call time, silence time, consultation text and voice waveform characteristics between the customer and the consultant. The voice waveform characteristics may include the height, width and section of the waveform. The speech waveform characteristics are used for emotional analysis. Sensibility analysis will be described later.

The consultation quality evaluation module 20 analyzes the evaluation data collected by the evaluation data collection module 10 and evaluates the consultation quality of the evaluation data by the predetermined evaluation items. The evaluation items are designed to evaluate the quality of counseling. The evaluation items may include the level of business knowledge, call duration, reason for misconduct, customer orientation, and consultation type.

The consultation quality evaluation module 20 includes a classification unit 21 and a consultation quality evaluation unit 22. [

The classification unit 21 classifies the evaluation data collected by the evaluation data collection module 10 according to a predetermined business domain.

The service area may include goods operated by the agency, or services provided by the agency to the customer.

For example, if the institution is a financial institution, the products operated by the institution may include deposits, withdrawals, loans, internet banking, cards, bancassurance and insurance. And other services required for operation. Here, the products operated by financial institutions and the services provided to customers are not specifically distinguished. The scope of work can be divided into various categories according to the work areas and work characteristics of financial institutions.

The consultation quality evaluation unit 22 evaluates the consultation quality of the evaluation data according to evaluation items according to the evaluation reference model after detecting a predetermined evaluation reference model according to the business area classified by the classification unit 21. The consultation quality evaluation unit 22 scales the evaluation results of each evaluation item so that the manager or the like can intuitively recognize the consultation quality evaluation result.

The evaluation criterion model is a pre-set model for evaluating the consultation quality and can be set for each evaluation item.

The consultation quality evaluation unit 22 classifies each evaluation data according to the business area classified by the classification unit 21 and classifies the evaluation data into evaluation items such as business knowledge level, And consultation type.

The consultation quality evaluation unit 22 includes a business knowledge level evaluation unit 221, a talk time analyzing unit 222, a consultation reason analyzing unit 223, a customer propensity classifying unit 224, .

The business knowledge level evaluation unit 221 compares the evaluation data with the business knowledge evaluation model of the evaluation reference model corresponding to the business area, and evaluates the business knowledge level of the consultant according to the comparison result. That is, the business knowledge level evaluation unit 221 evaluates the business knowledge level of the consultant by analyzing the similarity of the evaluation data with the business knowledge evaluation model of the evaluation reference model corresponding to the business domain.

The knowledge evaluation model is a reference model for evaluating the level of knowledge of counselors. The knowledge assessment model may include a counseling script, a Frequently Asked Questions (FAQ), and a counseling information. The counseling script may not only be predefined by the institution, but may also include a counseling script of the counselor. That is, the business knowledge level evaluator 221 can evaluate the level of business knowledge by analyzing the accuracy of the consultant's answer to the query items of the customer.

In addition, the business knowledge level evaluation unit 221 evaluates the level of business knowledge of the counselor based on the emotional analysis result of the evaluation data. In this case, the business knowledge level evaluation unit 221 uses the STT data of word overlap, silence, speed of speech, affirmative negative keyword, first language, first language, affirmative word, Analyze. The STT data may include speech waveform characteristics, for example, the height, width, and duration of the speech waveform.

That is, the business knowledge level evaluation unit 221 analyzes the accuracy of the consultant's answer based on the similarity of the evaluation data with the business knowledge evaluation model of the evaluation reference model corresponding to the business area, and analyzes the voice waveform characteristics of the STT data of the evaluation data To evaluate the level of counselor's work knowledge by evaluating the counselor's kindness.

In addition, the business knowledge level evaluation unit 221 warns that the evaluation level of business information can not be evaluated when the noise level of the evaluation data is equal to or higher than the setting level or the evaluation data is out of the business area.

In the case of a consultation call, it is possible that normal noise can not be provided due to excessive noise in the surrounding area, or a consultation on an area other than the work area in which the consultant can not be consulted may be requested.

Accordingly, the business knowledge level evaluation unit 221 determines that it is impossible to evaluate the business knowledge level of the evaluation data when the noise level of the STT data of the evaluation data is equal to or higher than the set level or the evaluation data deviates from the business scope. In this case, the business knowledge level evaluation unit 221 outputs the business knowledge level evaluation disabled through the output unit 30 or transmits the business knowledge level evaluation disabled to the administrator terminal 40. Accordingly, the manager can control the business knowledge level evaluation unit 221 through the administrator terminal 40 to exclude the evaluation data from the evaluation object.

The talk time analyzing unit 222 compares the evaluation data with the talk time evaluation model of the evaluation reference model corresponding to the business area and analyzes the talk time of the consultant according to the comparison result, Analyze silence interval and reason for silence.

Here, the talk time evaluation model is a reference model set in advance for analyzing the silence period and reason for silence.

Normally, the counselor can check other manuals to find answers to the inquiries of the customer during the consultation, or press the silent button for personal reasons to prevent the voice of the consultant from being delivered to the customer.

Accordingly, the talk time analyzing unit 222 extracts the silence time from the evaluation data through the time interval between the customer question and the response time, and determines whether the extracted silence time is more than the set silence time of the talk time evaluation model, The reason for the silence of the consultant is analyzed based on whether or not it matches the predetermined setting pattern.

In this case, the talk time analyzing unit 222 can analyze the talk time only for the evaluation data whose talk time of the evaluation data is equal to or longer than the predetermined talk time.

On the other hand, if the noise level of the evaluation data is equal to or higher than the set level or the evaluation data is out of the business area, the talk time analyzing unit 222 determines that the talk time analysis is impossible for the evaluation data through the output unit 30. In this case, the talk time analyzing unit 222 outputs the talk time analysis impossible through the output unit 30 or transmits the talk time analysis impossible to the administrator terminal 40. Accordingly, the manager can control the talk time analyzing unit 222 through the manager terminal 40 to exclude the evaluation data from the evaluation object.

The Ocontact Reasoning Analysis Unit 223 compares the evaluation data with the Ocontact Analysis Model of the evaluation reference model corresponding to the job domain and analyzes the reasons for the Ocontact consultation according to the comparison result. That is, the O-consulting reason analyzing unit 223 compares the evaluation data with the similarity of the O-consulting analysis model corresponding to the job domain and analyzes the reasons for the O-consulting of the consultant according to the comparison result.

Here, the O Consultation Analysis Model is a reference model for analyzing the reasons for the O consulting of the counselor. The O consultation analysis model may include counseling scripts, Frequently Asked Questions (FAQ), and misexcretation information. The counseling script may not only be predefined by the institution, but may also include a counseling script of the counselor.

On the other hand, when the noise level of the evaluation data is equal to or higher than the set level or the evaluation data is out of the business area, the O-consulting reason analyzing unit 223 determines that the O-consulting analysis is impossible through the output unit 30. In this case, the O-consulting reason analyzing unit 223 outputs the O-consulting reason analysis impossible report via the output unit 30 or transmits the O-consulting reason analysis impossible report to the manager terminal 40. Accordingly, the manager can control the false-consulting reason analyzing unit 223 through the administrator terminal 40 to exclude the evaluation data from the evaluation object.

The customer propensity classifier 224 compares the evaluation data with the emotional analysis model of the evaluation criterion model corresponding to the business domain and classifies the customer propensity into positive tendency or negative tendency according to the comparison result. Here, emotional analysis model is a reference model for classifying customer propensity. The emotional analysis model may include a positive keyword indicating that the customer is positive and a negative keyword indicating a negative tendency of the customer.

Accordingly, the customer propensity classifying section 224 classifies the customer propensity based on whether the positive or negative keyword exists in the evaluation data, and the number of repetitions of the positive keyword or the negative keyword.

On the other hand, if the noise level of the evaluation data is equal to or higher than the set level or the evaluation data is out of the business area, the customer propensity classifying unit 224 determines that the customer tendency classification is impossible for the evaluation data through the output unit 30. In this case, the customer propensity classifying unit 224 outputs the customer propensity class not possible through the output unit 30 or transmits the customer propensity class not to the administrator terminal 40. Accordingly, the manager can control the customer propensity classifying section 224 through the manager terminal 40 to exclude the evaluation data from the evaluation object.

The consultation type classifying unit 225 classifies the post-treatment classification details of the consultation type based on the evaluation data. That is, the consultation type classifier 225 analyzes the evaluation data to generate a post-processing classification history model of the evaluation reference model, and classifies the consultation type of the evaluation data based on the analysis result. In addition, the consultation type classifier 225 analyzes the evaluation data and generates a post-processing classification history model of the evaluation criteria model. Then, the post-treatment classification history model of the generated evaluation criteria model and the post- So that the consultant can confirm whether or not the post-processing classification details are correctly inputted.

On the other hand, if the noise level of the evaluation data is equal to or higher than the set level or the evaluation data is out of the business area, the consultation type classifying unit 225 determines that the consultation type classification is impossible for the evaluation data through the output unit 30. In this case, the consultation type classification unit 225 outputs the consultation type classification impossibility through the output unit 30 or transmits the consultation type classification impossibility to the administrator terminal 40. Accordingly, the administrator can control the consultation type classification unit 225 through the administrator terminal 40 to exclude the evaluation data from the evaluation target.

The output unit 30 outputs the consultation quality evaluation result of the consultation quality evaluation module 20. That is, the output unit 30 includes a business knowledge level evaluation unit 221, a talk time analyzing unit 222, an O consulting reason analyzing unit 223, a customer propensity classifying unit 224, and a consultation type classifying unit 225 The output is outputted in various formats such as image, text, graph, table, animation, etc., and these outputs are statistically processed for each counselor, time period and evaluation item, Or to the administrator terminal 40.

The output unit 30 includes an output management unit 31 and a statistical processing unit 32.

The statistical processing unit 32 is provided with an evaluation unit 223 for evaluating the service from each of the business knowledge level evaluation unit 221, the talk time analyzing unit 222, the O consulting reason analyzing unit 223, the customer propensity classifying unit 224, The results are input and statistically processed to generate statistical analysis results for each counselor, the organization to which the counselor belongs, the time frame and the evaluation items.

The output management unit 31 outputs the outputs of the business knowledge level evaluation unit 221, the talk time analyzing unit 222, the ocontact reason analyzing unit 223, the customer propensity classifying unit 224 and the consultation type classifying unit 225 And outputs at least one of image, text, graph, table, and animation in a combined format. In this case, the output management unit 31 can output the processing result of the statistical processing unit 32 in combination with images, texts, graphs, tables, animations, and the like, and can output the business knowledge level evaluation unit 221, The consultation reason analyzing unit 223, the customer propensity classifying unit 224, and the consultation type classifying unit 225, for each business domain, the counselor, and the time zone, respectively.

Also, the output management unit 31 can monitor and output the consultation quality evaluation result in real time as described above.

The manager terminal 40 includes the above-described business knowledge level evaluation unit 221, talk time analyzing unit 222, ocontact reason analyzing unit 223, customer propensity classifying unit 224 and consultation type classifying unit 225 If the noise level of the evaluation data is equal to or higher than the setting level or the evaluation data is out of the business area and the consultation type classification is impossible, the user is connected to the consultation quality evaluation module 20 according to the control command of the manager, It is excluded from evaluation.

The administrator terminal 40 may include a personal computer (PC), a laptop computer, a smart terminal, and the like.

Hereinafter, a consultation quality inherent evaluation method according to an embodiment of the present invention will be described with reference to FIG.

2 is a flowchart of a consultation quality inheritance evaluation method according to an embodiment of the present invention.

Referring to FIG. 2, the evaluation data collection module 10 receives evaluation data to be consultation quality evaluation objects from an agent terminal or the like (S10).

The evaluation data includes STT data that is converted into Speech To Text (STT) data of a post-processing classification made by the consultant after consultation between the client and the consultant, and a consultation call between the client and the consultant.

As the evaluation data are collected, the classification unit 21 classifies the evaluation data collected by the evaluation data collection module 10 into a predetermined business area (S20).

If the evaluation data is classified by the classification unit 21, the business knowledge level evaluation unit 221 compares the evaluation data with the business knowledge evaluation model of the evaluation reference model corresponding to the business area, . That is, the business knowledge level evaluation unit 221 evaluates the business knowledge level of the consultant by analyzing the similarity of the evaluation data with the business knowledge evaluation model of the evaluation reference model corresponding to the business domain.

In addition, the business knowledge level evaluation unit 221 evaluates the level of business knowledge of the counselor based on the emotional analysis result of the evaluation data. In this case, the business knowledge level evaluation unit 221 uses the STT data of word overlap, silence, speed of speech, affirmative negative keyword, first language, first language, affirmative word, Analyze.

In this manner, the business knowledge level evaluation unit 221 analyzes the similarity of the evaluation data with the business knowledge evaluation model of the evaluation reference model corresponding to the business area, and evaluates the business knowledge level of the consultant based on the emotion analysis result of the evaluation data (S30).

The talk time analyzing unit 222 compares the evaluation data with the talk time evaluation model of the evaluation reference model corresponding to the business area, and analyzes the talk time of the consultant according to the comparison result (S40). In this case, the talk-time analyzing unit 222 analyzes the silence period and reason for silence that the consultant does not correspond to the customer's question. That is, the talk time analyzing unit 222 extracts the silence time from the evaluation data through the time interval between the customer's question and the response time, and if the extracted silence time is more than the setting silence time of the talk time evaluation model, The reason for the silence of the consultant is analyzed based on whether or not it matches the predetermined setting pattern.

The O-consulting reason analysis unit 223 compares the evaluation data with the O-consultation analysis model of the evaluation reference model corresponding to the business area and analyzes the O-consulting reason of the consultant according to the comparison result (S50). In this case, the O-consulting reason analyzing unit 223 compares the similarity of the evaluation data with the O-consulting analysis model corresponding to the business area, and analyzes the reasons for the O-consulting of the consultant according to the comparison result.

The customer propensity classifier 224 compares the evaluation data with the emotion analysis model of the evaluation criterion model corresponding to the business domain, and classifies the customer propensity into either positive propensity or negative propensity according to the comparison result (S60). In this case, the customer propensity classifying section 224 classifies the customer propensity based on whether the positive or negative keyword exists in the evaluation data, and the number of iterations of the positive keyword or the negative keyword.

The consultation type classification unit 225 classifies the post-processing classification details of the consultation type based on the evaluation data (S70). In this case, the consultation type classification unit 225 analyzes the evaluation data to generate a post-processing classification history model of the evaluation reference model, and after the consultant calls the post-processing classification history model of the evaluation reference model, Compare the results with the classification details, and check whether the results of the comparison are consistent with each other, so that the counselor can confirm whether or not the post-treatment classification details are accurately described.

On the other hand, in the evaluation process of the business knowledge level evaluating unit 221, the talk time analyzing unit 222, the consultation reason analyzing unit 223, the customer propensity classifying unit 224 and the consultation type classifying unit 225, The manager terminal 40 accesses the consultation quality evaluation module 20 in accordance with the control command of the manager and transmits the evaluation data to the consultation quality evaluation module 20 It can be excluded from the evaluation subject.

The output unit 30 outputs the outputs of the business knowledge level evaluation unit 221, the talk time analyzing unit 222, the O consulting reason analyzing unit 223, the customer propensity classifying unit 224 and the consultation type classifying unit 225 (S80). The output of the statistical processing is statistically processed by the counselor, by time, and by the evaluation item, and the result of the statistical analysis is output to a display (not shown) , Or the administrator terminal (40).

Next, an example of a screen for outputting the consultation quality evaluation result will be described with reference to Figs. 3 to 9. Fig.

FIG. 3 is a view showing a screen for real-time monitoring status of consultation quality evaluation results according to an embodiment of the present invention, FIG. 4 is a detail view of consultation quality analysis according to an embodiment of the present invention, FIG. 6 is a view showing an evaluation management screen according to an embodiment of the present invention, and FIG. 7 is a view showing an evaluation management screen according to an embodiment of the present invention. FIG. 8 is a view showing a screen for evaluation statistics according to a retrieval condition according to an embodiment of the present invention, and FIG. 9 is a view showing an example of the evaluation condition according to an embodiment of the present invention. FIG. 3 is a diagram showing a screen for the consultation type transition and the consulting general index according to the present invention.

FIG. 3 is a screen showing a result of monitoring the consultation quality evaluation result in real time. FIG. 3 shows the number and type of objects for which consultation quality was evaluated, the number of objects for which consultation quality evaluation was impossible, the evaluation results, and the time-based variation thereof.

4 is a detailed view of the consultation quality evaluation result during a specific inquiry period. FIG. 4 shows evaluation data to be analyzed during a specific inquiry period and their consultation quality evaluation results.

FIG. 5 is a detailed view of consultation quality evaluation results for a specific consultant during a specific inquiry period. In FIG. 5, consultation quality evaluation results are shown for each of a high score counselor and a low score counselor.

FIG. 6 is a screen showing in detail the consultation quality evaluation result for a specific consultant. In Fig. 6, the consulting quality is evaluated in detail for the consultant, and as a result, the consultation quality evaluation result is shown in detail.

7 is a screen for a query condition and an evaluation history list. In Fig. 7, an inquiry item for inputting the inquiry condition for inquiring the consultation quality evaluation result in detail, and an evaluation item matched to the inquiry condition are shown.

FIG. 8 is a screen for evaluation statistics according to a query condition. FIG. 8 shows the result of statistical processing on the consultation quality evaluation result matched to the specific inquiry condition.

9 is a screen for the consultation type transition and the counseling general index. In FIG. 9, the results of the statistical processing on the consultation quality evaluation results are shown by graphs, images, and characters.

As described above, the consultation quality inherent evaluation apparatus according to the embodiment of the present invention evaluates the consultation quality of the entire consultation data by comparing the consultation data that the consultant has made with the customer to the pre-stored evaluation reference model, The quality of counseling for counseling materials, such as counseling and emotional evaluation, can be evaluated in various aspects, and can be utilized in training planning and education for counselors.

In addition, the consultation quality transfer evaluation apparatus according to another embodiment of the present invention classifies the customer orientation on the basis of the sensitivity evaluation so that the consultation service for the customer can be performed more efficiently.

In addition, the consultation quality transfer evaluation device according to another embodiment of the present invention provides the consultation quality evaluation result and the business status in real time, so that the management of the consultation quality can be performed in real time.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is clearly understood that the same is by way of illustration and example only and is not to be taken by way of limitation, I will understand. Accordingly, the true scope of the present invention should be determined by the following claims.

10: Evaluation data collection module 11: First evaluation data collection part
12: second evaluation data collection section 20: consultation quality evaluation module
21: classification section 22: consultation quality evaluation section
221: Business knowledge level evaluation unit 222: Talk time analysis unit
223: Reason for Consultation Analysis Unit 224: Customer Propensity Classification Unit
225: consultation type classifying section 30: output section
31: output management unit 32: statistical processing unit
40:

Claims (18)

An evaluation data collection module for collecting assessment data to be subjected to consultation quality evaluation;
A consultation quality evaluation module for analyzing the evaluation data collected by the evaluation data collection module and evaluating the consultation quality for each predetermined evaluation item; And
And an output unit for outputting a consultation quality evaluation result evaluated by the consultation quality evaluation module,
Wherein the consultation quality evaluation module comprises: a classification part for classifying all counseling calls into a predetermined business area; And a consultation quality evaluating unit for evaluating a consulting quality of the evaluation data according to the evaluation items according to the evaluation reference model after detecting a predetermined evaluation reference model according to the business area classified in the classification unit,
Wherein the consultation quality evaluation unit compares the evaluation data with a business knowledge evaluation model of the evaluation reference model corresponding to the business area and evaluates a business knowledge level of the consultant according to the comparison result; A talk time analyzer for comparing the evaluation data with the talk time evaluation model of the evaluation reference model corresponding to the business area and analyzing the talk time of the consultant according to the comparison result; An o consultation reason analyzer for comparing the evaluation data with a misclassification analysis model of the evaluation criterion model corresponding to the work area and analyzing a consultant's reason for misconduct according to the comparison result; A customer propensity classifying unit for comparing the evaluation data with a sensitivity analysis model of the evaluation reference model corresponding to the business area and classifying the customer propensity according to the comparison result; And a consultation type classification unit for comparing the post-treatment classification history model of the evaluation reference model with the consultation type and the post-treatment classification details directly created by the consultant after the consultation call and classifying the consultation type according to the comparison result Quality transfer evaluation device.
2. The apparatus of claim 1, wherein the evaluation data collection module
A first evaluation data collection unit for collecting post processing classification data created by the consultant after consultation between the client and the consultant; And
And a second evaluation data collection unit for collecting STT (Speech To Text) data obtained by converting a consultation call between the customer and the consultant into text.
3. The apparatus according to claim 2, wherein the STT data includes at least one of talk time, silence time, consultation text, and voice waveform characteristics.
delete 2. The system of claim 1,
A service operated by an organization, and a service item provided by an organization to a customer.
delete The system according to claim 1, wherein the business knowledge level evaluation unit
Analyzing the similarity between the evaluation data and the business knowledge evaluation model of the evaluation reference model corresponding to the business area and evaluating the business knowledge level of the consultant based on the emotion analysis result of the evaluation data; Evaluation device.
8. The method of claim 1 or 7, wherein the task knowledge evaluation model of the evaluation criterion model includes at least one of a predefined consultation script, a frequently asked questions (FAQ) Evaluation device.
The system according to claim 1, wherein the business knowledge level evaluation unit
And notifies that the evaluation of the business knowledge level is impossible with respect to the evaluation data if the noise level of the evaluation data is equal to or higher than the set level or the evaluation data deviates from the business area.
The apparatus of claim 1, wherein the talk time analyzing unit
And a step of determining whether or not the extracted silence time is equal to or longer than the set silence time of the talk time evaluation model or whether the answer content matches the preset setting pattern And analyzing the cause of the silence of the consultant through the questionnaire.
The apparatus of claim 10, wherein the talk time analyzing unit
And notifies that the talk time analysis can not be performed on the evaluation data when the noise level of the evaluation data is equal to or higher than the setting level or when the evaluation data is out of the business area.
The method according to claim 1,
Wherein the evaluation data is compared with the similarity of the o consultation analysis model corresponding to the business area, and a reason for consulting the counselor based on the comparison result is analyzed. 13. The method according to claim 12,
Wherein when the noise level of the evaluation data is equal to or higher than the setting level or when the evaluation data deviates from the business area, the evaluating unit warns the user that the analysis of the consulting cause is impossible.
The system of claim 1, wherein the customer preference classifier
Wherein the customer's tendency is classified into one of a positive tendency and a negative tendency based on the emotional analysis result of the evaluation data.
15. The system of claim 14, wherein the customer preference classifier
And notifies the customer that the customer is not able to classify the evaluation data if the noise level of the evaluation data is equal to or higher than the set level or the evaluation data is out of the business area.
2. The apparatus according to claim 1, wherein the consultation type classification unit
Processing classification history model of the evaluation criterion model for the consultation type and the post-treatment classification details directly generated by the consultant after the consultation call and evaluating whether the consultant has correctly input the post-treatment classification details according to the comparison result Consultation quality transfer evaluation device.
17. The method according to claim 16, wherein the consultation type classification unit
And notifies that the consultation type can not be classified for the evaluation data if the noise level of the evaluation data is equal to or higher than the setting level or the evaluation data is out of the business area.
The system according to claim 1, wherein the output unit outputs the output of the business knowledge level evaluation unit, the talk time analyzing unit, the ocontact reason analyzing unit, the customer propensity classifying unit, and the consultation type classifying unit, And a statistical processing unit for statistically processing the statistical information.

Images