WO2018015986A1

WO2018015986A1 - System, method, and program for classifying customer's assessment data, and recording medium therefor

Info

Publication number: WO2018015986A1
Application number: PCT/JP2016/003439
Authority: WO
Inventors: 斐周
Original assignee: 株式会社Ｆｒｏｎｔｅｏ
Priority date: 2016-07-22
Filing date: 2016-07-22
Publication date: 2018-01-25

Abstract

This system for classifying customer's assessment data is provided with: a data acquisition unit which extracts assessment factors, both from data known to be improvement request data and from data known to be usage termination data; an assessment factor storage unit which stores the assessment factors acquired by the data acquisition unit; an assessment factor extraction unit which extracts, from customer's assessment data to be classified, determination factors corresponding to assessment factors stored in the assessment factor storage unit; and a classification determination unit which determines, on the basis of the determination factors extracted by the assessment factor extraction unit, whether or not the customer's assessment data is improvement request data, and if it is determined that the customer's assessment data is improvement request data, classifies the customer's assessment data as improvement request data, whereas if it is determined that the customer's assessment data is not improvement request data, the classification determination unit determines whether or not the customer's assessment data is usage termination data, and if it is determined that the customer's assessment data is usage termination data, the classification determination unit classifies the customer's assessment data as usage termination data.

Description

CUSTOMER EVALUATION DATA CLASSIFICATION SYSTEM, METHOD, PROGRAM, AND ITS RECORDING MEDIUM

The present invention relates to a customer evaluation data classification system, method, program, and recording medium for classifying customer evaluation data (for example, document or audio data).

Generally, a service provider discloses contact information for services provided by the service provider. The customer who receives the service of the service provider evaluates the received service based on the evaluation data. For example, the data includes documents such as letters and e-mails, or voice data such as voice messages. Hereinafter, in this specification, a document will be described as a representative example of evaluation data. The evaluation includes a positive evaluation such as praise for the service and a negative evaluation such as a claim for the service. Since negative evaluation includes problems with services, analyzing this negative evaluation leads to improvement of the quality of services provided.

For example, Patent Document 1 extracts related words related to user dissatisfaction from data related to user dissatisfaction, and extracts dissatisfaction that can be connected to other services from the expression of dissatisfaction in the related words. The technology to do is disclosed.

JP 2010-204865 A

In Japanese Patent Laid-Open No. 2004-26883, when a related word related to user dissatisfaction is extracted from a set of test data and a dissatisfaction that can be connected to other services from the expression of dissatisfaction in the related word is determined, the degree of dissatisfaction is determined. Relative criteria for other data belonging to the set are classified as “Negative”, “Neutral”, and “Affirmative”. However, with this technique, it is not possible to classify and extract customer dissatisfaction based on absolute criteria, rather than relative criteria within test data. That is, it is unclear whether or not the extracted customer dissatisfaction is an obstacle to maintaining the trust of the service provider. It is required to classify and extract customer dissatisfaction on an absolute basis.

A customer evaluation data classification system that classifies customer evaluation data evaluated by a customer, and evaluates each of data that is known to be improvement request data and data that is known to be use stop data. A data acquisition unit for extracting components, an evaluation component storage unit for storing evaluation components acquired by the data acquisition unit, and an evaluation configuration stored in the evaluation component storage unit from customer evaluation data to be classified Based on the evaluation component extraction unit that extracts the determination component corresponding to the element and the determination component extracted by the evaluation component extraction unit, it is determined whether or not the customer evaluation data is improvement request data. When it is determined that the data is data, the customer evaluation data is classified as improvement request data, and when the customer evaluation data is determined not to be improvement request data Determines whether the customer evaluation data is use stop data, and if it is determined that the customer evaluation data is use stop data, the customer includes a classification determination unit that classifies the customer evaluation data as use stop data. Solve by evaluation data classification system.

Customer evaluation data can be classified into “Improvement request data”, “Dissatisfaction data”, and “Use stop data”.

It is the figure which showed the principle of the customer evaluation data classification system of this invention. It is a figure of the hardware constitutions of the customer evaluation data classification system of this invention. It is a functional block diagram of the customer evaluation data classification system of the present invention. It is the figure which showed the algorithm of the program of the customer evaluation data classification system of this invention.

[Data format processed by customer evaluation data classification system]
In the present embodiment, “data” may be any data expressed in a format that can be processed by a computer. The data may be, for example, unstructured data whose structure definition is incomplete at least in part, and document data (for example, e-mail (attached file header) Information), technical documents (including a wide range of documents explaining technical matters such as academic papers, patent publications, product specifications, design drawings, etc.), presentation materials, spreadsheets, financial statements, meeting materials, Record reports, sales documents, contracts, organization charts, business plans, company analysis information, electronic medical records, web pages, blogs, comments posted on social network services, etc., audio data (eg conversation / music) Data), image data (eg, data composed of a plurality of pixels or vector information), video data (eg, Broadly includes data formed) including a plurality of frame images (not limited to these examples).

[Principle of Customer Evaluation Document Classification System]
Hereinafter, as this embodiment, the customer evaluation data classification system, method, program, and recording medium thereof will be described as a customer evaluation document classification system based on document data as a representative example of data. Here, the principle of classifying customer evaluation documents according to the degree of evaluation in the customer evaluation document classification system of the present invention will be described with reference to FIG. FIG. 1 is a diagram illustrating the principle of classifying documents according to customer evaluations in a customer evaluation document classification system. The evaluation document evaluated by the customer with respect to the service provider can be classified into “improvement request document 51”, “usage stop document 52” or “dissatisfaction document 53” that declares the service provider to stop using it.

Here, in this specification, these documents are defined as follows. “Improvement request document 51” is that the customer is dissatisfied with the service provider (operator) for providing the service (or product), but the expectation for improvement of the service is preferable. A document containing opinions. In other words, the improvement request document is a document including an expression that encourages the operator to improve the product or service based on dissatisfaction with the product or service provided by the operator. For example, if a word that expresses favor or loyalty to the product or service, such as “improvement”, “fun”, “review”, “review”, “fan”, “favorite”, etc., It may correspond to “Improvement Request Document”. Or, even if the above word is not included, if there is a statement stating that the product or service has been frequently purchased or used (or used) before, The document may correspond to an “improvement request document”.

On the other hand, the “usage suspension document 52” includes an opinion that the customer is dissatisfied with the service provider (operator) to provide the service (or product) and does not use the product or service. Is a document. For example, if a negative word for the product or service is included, such as “I feel bad”, “I ca n’t”, “I do n’t like it”, “I do n’t feel”, It may fall under “document”. Or, even if the above word is not included, if there is a clear statement that the product or service will not be used again (for example, “Never use it again”), the document It may correspond to “suspended document”.

It should be noted that a document that is neither an improvement request document nor a use suspension document and that the customer expresses dissatisfaction with the product or service is the “dissatisfaction document 53”.

First, the learning process is performed. In the learning process, sample evaluation documents are collected as learning data that are known to belong to any one of the improvement request document 51, the use stop document 52, and the dissatisfaction document 53. Among these, firstly, “improvement request constituent elements composed of constituent elements that are meant to affirm the service provider” are extracted from the improvement request document 51. Then, the “usage stop component which means that the service provider service is not used” is extracted from the use stop document 52. The extracted element is stored in the storage device.

Subsequently, the evaluation process is performed. A component constituting the document is extracted from the test evaluation document, and it is determined whether the component corresponds to the improvement request component stored in the storage device. At this time, whether or not it is applicable is determined to be applicable if the score value of the test evaluation document is calculated and the score value is higher than a predetermined threshold value. If it falls under the improvement request component, it is determined that the test evaluation document is the improvement request document 51.

When the test evaluation document is not the improvement request document 51, it is subsequently determined whether or not the component extracted from the test evaluation document corresponds to the use suspension component. If it falls under the use stop component, it is determined that the test evaluation document is the use stop document 52.

If the test evaluation document is neither the improvement request document 51 nor the use suspension document 52, the document is determined as a dissatisfied document 53. If the test evaluation document belongs to both the improvement request document 51 and the use stop document 52, it is determined as the improvement request document 51.

The improvement request component is, for example, affirming the facts about the adjectives and numerators + usages that represent “the number of (owned products) / (used)” from the sample document of improvement requests. ”Verb”. For example, “I have a lot of items from your store” or “I went to the store every week”. The decommissioning component is a negative expression or word for the brand and a clear expression that the store or service is not reused.

[Hardware configuration of customer evaluation document classification system]
With reference to FIG. 2, a customer evaluation document classification system (hereinafter simply referred to as “system”) of the present invention will be described. FIG. 2 is an example of a hardware configuration of the system 1. The system 1 includes a server device 10 and a client terminal 11. The server device 10 includes an arithmetic device 10a that performs calculation and a storage device 10b that stores data.

The customer evaluation document classification system 1 includes a server device 10 and a client terminal 11. The server device 10 includes an arithmetic device 10a that performs calculation and a storage device 10b that stores data. The server device 10 can execute main processing of data analysis. The client terminal 11 can execute a data analysis related process in the server device 10. The storage device 10b is, for example, any recording medium (for example, a memory or a hard disk) that can store data (including digital data and analog data). The arithmetic device 10a is a controller (for example, a central processing unit (CPU)) that can execute a control program stored in a recording medium. The computing device 10a is a computer or a computer system (a system that realizes data analysis by operating a plurality of computers in an integrated manner) that analyzes data stored at least temporarily in a recording medium. The computing device 10a may be configured as a management computer (not shown) in the form of an external device of the server device 10, and the storage device 10b is configured as the data storage server device 13 of the external storage device of the server device 10. You may make it comprise with a form.

The management computer (not shown) may include, for example, a memory, a controller, a bus, an input / output interface, and a communication interface. Note that a customer evaluation morpheme classification program is stored in a memory serving as a recording medium included in each of the server device 10 and the client terminal 11. By executing the customer evaluation morpheme classification program, the application program (software resource) and the hardware resource cooperate to operate each device.

The storage device 10b is composed of, for example, a disk array system, and can include a database that records data and results of evaluation / classification of the data. The server device 10 and the storage device 10b are connected by a direct connection method (DAS) or a storage device area network (SAN).

The client terminal 11 presents data in the middle of the processing process in the server device 10 to the user. As a result, the user can input, that is, provide classification information through bidirectional exchange via the client terminal 11. The client terminal 11 includes, for example, a memory, a controller, a bus, an input / output interface (for example, a keyboard, a display, etc.), and a communication interface (communication means using a predetermined network). For communication). The client terminal 11 may be configured to include an input device 12 such as a scanner.

Note that the hardware configuration shown in FIG. 2 is merely an example, and the system 1 can be realized by other hardware configurations. For example, a configuration in which part or all of all the processes are executed in the server device 10 may be used, or a part or all of the processing may be executed in the client terminal 11. In this embodiment, the input device 12 is connected to the client terminal 11 and can transmit to the server device 10. However, the input device 12 directly connects to the server device 10 and inputs data to the server from here. May be. It will be understood by those skilled in the art that various hardware configurations capable of realizing the system 1 exist, and the configuration is not limited to the configuration illustrated in FIG. 2, for example.

[Functional block configuration of customer evaluation document classification system]
A functional block diagram of the customer evaluation document classification system of the present invention will be described with reference to FIG. FIG. 3 shows a functional block diagram of the customer evaluation document classification system 1. The customer evaluation document classification system 1 includes a data acquisition unit 21, an evaluation component acquisition unit 22, an evaluation component extraction unit 23, an evaluation component storage unit 24, and a classification determination unit 25.

The data acquisition unit 21 is a part that acquires a plurality of components that constitute a sample evaluation document of customer evaluation for a certain service provider. The component consists of a plurality of parts of speech and consists of a component requesting improvement that is affirmative to the service provider and a suspension of use that is a component that indicates that the service of the certain service provider is not used. It is classified as a component.

The evaluation component storage unit 24 stores the component acquired by the data acquisition unit 21. The classification determination unit 25 determines that the component of the test evaluation document is an improvement request document if the component is an improvement request component, and determines that the component is a use stop document if the component is a use stop component. If not, it is determined that the document is dissatisfied.

This function block functions as follows. First, in a learning process for a sample document, an “evaluation component” to be evaluated is input to the evaluation component acquisition unit 22. An evaluation component is a basic morpheme that can be identified semantically from the component. That is, one evaluation component is associated with a plurality of components that can be identified semantically. For example, “Quality” and “Quality” are components, respectively. “Quality” is selected as the evaluation component, and “Quality” and “Quality” A weight is assigned and associated between the “good quality” of the component and the evaluation component. As the “evaluation component”, “a component that is meant to affirm the service provider” that is a feature of the improvement request document 51 and “a service provider's service is not used” that is a feature of the use suspension document 52 Is a use suspension component that means "

The inputted “evaluation component” is stored in the evaluation component storage unit 24 (storage device 10b). In the evaluation process, the data acquisition unit 21 captures the document to be examined. Then, the document to be examined is finely divided into morphemes to create a component group of the document to be examined. The evaluation component extraction unit 23 extracts “determination component” corresponding to “evaluation component” stored in the evaluation component storage unit 24 from the component group of the document to be examined. The determination component is a morpheme corresponding to the evaluation component, and is a morpheme component associated with the evaluation component. For example, “quality is high” and “quality is high” corresponding to “quality is good” which is an evaluation component stored in the evaluation component storage unit 24 are extracted as determination components. The classification determination unit 25 performs classification based on the extracted “determination component” based on the weighting or the like previously associated with the “evaluation component”. As a result of the determination, it is determined and classified to which classification the document to be examined belongs, that is, the improvement request document 51 or the use stop document 52.

[Algorithm configuration of customer evaluation document classification system]
Next, the customer evaluation document classification algorithm will be described with reference to FIG. The same applies to the customer evaluation document classification method, program, and recording medium. The algorithm configuration of the customer evaluation document classification system is classified into a learning process based on learning data (S101 to S104) and an evaluation classification process (S105 to S108) for a document to be examined.

First, in the learning process, as learning data, a document that is known to be a “sample evaluation document of an improvement request document”, a document that is known to be a “sample document of a dissatisfied document”, and “ An evaluation component is extracted and selected from each document with a document known to be a “sample document” and determined (S101).

The selected evaluation component is fetched from a document known to be a “sample evaluation document for improvement request document” (S102). Similarly, an evaluation component is extracted also from a document that is known to be a “sample document for use stop document” (S103). The extracted evaluation components (“improvement request component” and “use stop component”) are stored in the storage unit. The extracted and selected evaluation components are stored in the storage device 10b. At this time, the degree to which the evaluation component contributes to the combination of the learning data and the classification information is calculated as the weight of the evaluation component, and the evaluation component and the weight are associated with each other and stored in the storage device 10b. It's okay.

Next, move on to the evaluation classification process. In the evaluation classification process, first, data of a test evaluation document is captured (S105). Then, a “determination component” corresponding to the evaluation component already stored in the storage unit and corresponding to the evaluation component is extracted from the document to be examined (S106). The extracted “determination component” is evaluated to determine whether the component is an “improvement request component”. The evaluation at this time can be determined using the score value as an index. The score value can be calculated by an arbitrary method based on the relationship between the “evaluation component” and the “determination component”. The calculation of the score value will be described later (S107). When the calculated score value exceeds a predetermined value as a threshold value, it is determined that the document is an “improvement request document” (S108). If the calculated score value does not exceed the predetermined value as the threshold value, it is determined using the score value as an index whether or not the extracted component is a “use stop component”. That is, when the score value calculated by an arbitrary method exceeds a predetermined value as a threshold value, it is determined that the document is a “use-stopped document”, and the calculated score value exceeds a predetermined value as a threshold value. If not, it is determined that all the documents are “unsatisfied documents” (S108).

As described above, the score value can be calculated by an arbitrary method. For example, various methods used in the field of machine learning or natural language processing (for example, a K-neighbor method, a method using a support vector machine, a method using a neural network, a method that assumes a statistical model for data (for example, , A method using a Gaussian process, etc.) and / or a method combining these, etc.), or based on various methods used in the field of statistics (eg, configuration) (Based on how often the element appears in the data). Here, the “determination component” may be partial data constituting at least part of the “evaluation component” of the sample document as sample data. For example, the morpheme, keyword, sentence, paragraph, And / or metadata (e.g., e-mail header information), partial audio that constitutes audio, volume (gain) information, and / or timbre information, partial images that constitute an image, partial pixels, And / or luminance information, or frame images, motion information, and / or three-dimensional information constituting a video.

When calculating the score value based on the frequency at which the evaluation component appears in the document to be examined, for example, the following calculation method can be considered. First, an evaluation configuration that constitutes each of a document that is known to be a “sample evaluation document for improvement request document” and a document that is known to be a “sample document for use stop document” as learning data Extract elements. Then, for example, a determination component having a high relationship with the evaluation component is extracted from the document to be examined. The determination component constitutes at least a part of the evaluation component. For example, an evaluation component corresponding to the determination component is referred to, and a weight associated with the evaluation component is referred to. In other words, each distribution in which the determination component appears according to the classification information is evaluated. For example, in the test document A, the appearance probability of the determination component when the test document A is assumed to be an improvement request document represents the degree of contribution to the improvement request as classification information for the determination component, and the test document A The appearance probability of the determination component when assuming that the document is a use stop document represents a degree of contribution to the use stop as classification information for the determination component. As a more specific example, by evaluating the component using the amount of transmitted information (for example, the amount of information calculated from a predetermined formula using the appearance probability of the determination component and the appearance probability of the classification information) The evaluation value of the component is calculated.

For example, the extraction and evaluation of the determination component may be continued (repeated) until the recall rate or the matching rate reaches a predetermined target value. In this case, the system may calculate the evaluation value wgt of the component using, for example, the following equation.

Here, wgt indicates the initial value of the evaluation value of the i-th component before evaluation. Wgt indicates the evaluation value of the i-th component after the Lth evaluation. γ means an evaluation parameter in the L-th evaluation, and θ means a threshold value in the evaluation. Thereby, for example, the larger the value of the calculated transmission information amount, the more the component can be evaluated as representing the characteristics of the predetermined classification information.

Next, an evaluation value is determined in association with the extracted component, and the component and the evaluation value are stored in the storage device 10b. Then, a component is extracted from the evaluation document, and an inquiry is made as to whether or not the component is stored in the storage device 10b. If the component is stored, an evaluation value associated with the component is stored in the storage device 10b. The evaluation data is evaluated based on the evaluation value. As a more specific example, a score value can be calculated by calculating the following expression using an evaluation value associated with a component constituting at least a part of the evaluation data.

m _j : frequency of occurrence of the i-th component
wgt _i : Evaluation value of the i-th component

[Processing after classification]
The customer evaluation document classification system can automatically suggest or execute preventive measures for the user according to the classification destination of the customer evaluation document. For example, “transfer the customer evaluation document to the product development department” may be defined as an action in the case of being classified as “improvement request document”. Thereby, for example, it is possible to improve the efficiency of complaint processing and product development. For example, when the product development department analyzes a customer evaluation document classified as “improvement request”, it becomes possible to enhance understanding of what a loyal customer wants. Further, for example, “transfer the customer evaluation document to the customer service center” may be defined as an action when the document is classified as “dissatisfied document”. As a result, for example, it is possible to expedite the return and exchange procedures.

[Example of implementation using software and hardware]
The control block of the above system may be realized by a logic circuit (hardware) formed in an integrated circuit (IC chip) or the like, or may be realized by software using a CPU. In the latter case, the system includes a CPU that executes a program (customer evaluation document classification program) that is software for realizing each function, and a ROM (Read CPU) in which the program and various data are recorded so as to be readable by the computer (or CPU). Only Memory) or a storage device (these are referred to as “recording media”), a RAM (Random Access Memory) for developing the program, and the like. And the objective of this invention is achieved when a computer (or CPU) reads the said program from the said recording medium and runs it. As the recording medium, a “non-temporary tangible medium” such as a tape, a disk, a card, a semiconductor memory, a programmable logic circuit, or the like can be used. The program may be supplied to the computer via an arbitrary transmission medium (such as a communication network or a broadcast wave) that can transmit the program. The present invention can also be realized in the form of a data signal embedded in a carrier wave in which the program is embodied by electronic transmission. Note that the above program can be implemented in any programming language. Also, any recording medium that records the above program falls within the scope of the present invention.

[Other Embodiments of the Invention]
The present invention can also be expressed as follows. That is, the data includes audio data. A data classification system for classifying a plurality of customer voice data, wherein each of the plurality of customer voice data includes at least a part of information related to customer evaluation of a product or service provided by an operator, and a processor; A memory for storing at least temporarily customer voice data, wherein the processor presents a portion of the plurality of customer voice data as reference data to a user via a display, and the classification input by the user The information is acquired, and the classification information is information for classifying the reference data by being associated with the presented reference data, and the reference component constituting at least a part of the reference data is acquired. For each classified information extracted from the reference data, A target component constituting at least a part of the data is extracted from the target data, and the target data constitutes another part of the plurality of customer voice data, and the data not associated with the classification information And an index is calculated using the reference component corresponding to the extracted target component, and the score is an index indicating the possibility of the target data being associated with predetermined classification information, The calculated index may be expressed as a data classification system that is presented to the user.

Further, the processor evaluates, as a weight, a degree to which the extracted reference component contributes to a combination of the reference data and classification information associated with the reference data, and the reference component and the reference data Associating the evaluated weights with each other, storing them in the memory, referring to the reference component corresponding to the extracted target component in the memory, and using the weight associated with the reference component, the index May be calculated.

Further, the processor can associate the classification information with the target data according to the score, execute an action for the customer according to the association, or present the action to the user.

The customer evaluation document classification system of the present invention is not limited to the above-described embodiments, and various modifications can be made within the scope of the claims, and the techniques disclosed in the different embodiments, respectively. Embodiments obtained by appropriately combining technical means are also included in the technical scope of the present invention. Furthermore, a new technical feature can be formed by combining the technical means disclosed in each embodiment.

DESCRIPTION OF SYMBOLS 1 Customer evaluation document classification system 10 Server apparatus 11 Client terminal 12 Input device 13 Data storage server apparatus 21 Data acquisition part 22 Evaluation component acquisition part 23 Evaluation component extraction part 24 Evaluation component storage part 25 Classification determination part 51 Improvement request document 52 Discontinued Document 53 Dissatisfied Document

Claims

A customer evaluation data classification system for classifying customer evaluation data evaluated by a customer,
A data acquisition unit that extracts an evaluation component from each of data that is known to be improvement request data and data that is known to be use suspension data;
An evaluation component storage unit that stores the evaluation component acquired by the data acquisition unit;
An evaluation component extraction unit that extracts a determination component corresponding to the evaluation component stored in the evaluation component storage unit from customer evaluation data to be classified;
Based on the determination component extracted by the evaluation component extraction unit, it is determined whether the customer evaluation data is improvement request data. If it is determined that the customer evaluation data is improvement request data, the customer evaluation data Is determined to be improvement request data, and when it is determined that the customer evaluation data is not improvement request data, it is determined whether or not the customer evaluation data is use stop data. A customer evaluation data classification system comprising: a classification determination unit that classifies the customer evaluation data as use stop data when determined.
The customer evaluation data classification system according to claim 1,
A customer evaluation data classification system for classifying that the customer evaluation data is dissatisfied data when it is determined that the customer evaluation data is not use stop data.
The customer evaluation data classification system according to claim 1 or 2,
The determination is a customer evaluation data classification system that calculates a score value based on a determination component and classification information, and classifies the customer evaluation data according to whether the score value exceeds a predetermined threshold.
A customer evaluation data classification method for classifying customer evaluation data evaluated by a customer,
Extracting an evaluation component from each of data that is known to be improvement request data and data that is known to be use stop data; and
Storing the evaluation component acquired by the data acquisition unit;
Extracting a judgment component corresponding to the evaluation component stored in the evaluation component storage unit from the customer evaluation data to be classified;
Based on the determination component extracted by the evaluation component extraction unit, it is determined whether the customer evaluation data is improvement request data. If it is determined that the customer evaluation data is improvement request data, the customer evaluation data Is determined to be improvement request data, and when it is determined that the customer evaluation data is not improvement request data, it is determined whether the customer evaluation morpheme is a use stop morpheme, and is use stop data. A customer evaluation data classification method including a step of classifying the customer evaluation data as use stop data when it is determined.
A customer evaluation data classification program that is executed on a computer and classifies customer evaluation data evaluated by a customer, the customer evaluation data classification program including:
An evaluation component is extracted from each of data that is known to be improvement request data and data that is known to be use stop data.
Storing the obtained evaluation component;
From the customer evaluation data to be classified, a determination component corresponding to the evaluation component stored in the evaluation component storage unit is extracted,
Based on the determination component extracted by the evaluation component extraction unit, it is determined whether the customer evaluation data is improvement request data. If it is determined that the customer evaluation data is improvement request data, the customer evaluation data Is determined to be improvement request data, and when it is determined that the customer evaluation data is not improvement request data, it is determined whether or not the customer evaluation data is use stop data. A customer evaluation data classification program for classifying that the customer evaluation data is use stop data when judged.
A recording medium that stores a customer evaluation data classification program that is executed by a computer and classifies customer evaluation data evaluated by a customer. The customer evaluation data classification program includes:
An evaluation component is extracted from each of data that is known to be improvement request data and data that is known to be use stop data.
Storing the obtained evaluation component;
From the customer evaluation data to be classified, a determination component corresponding to the evaluation component stored in the evaluation component storage unit is extracted,
Based on the determination component extracted by the evaluation component extraction unit, it is determined whether the customer evaluation data is improvement request data. If it is determined that the customer evaluation data is improvement request data, the customer evaluation data Is determined to be improvement request data, and when it is determined that the customer evaluation data is not improvement request data, it is determined whether or not the customer evaluation data is use stop data. A recording medium that classifies the customer evaluation data as use stop data when it is determined.