JP2019194793A - Information processing apparatus and program - Google Patents

Abstract

To enable a customer to grasp the possibility of continuous use for a certain service or product.SOLUTION: An expression classification unit 42 classifies transmission information acquired by a transmission information acquisition unit 31 for each type of factor representing a user's feeling about a service or product. A loyalty degree calculation unit 46 specifies a relationship or degree between a plurality of types of factors classified by the expression classification unit 42 from an attitude factor degree, a behavior factor degree, and a switching factor degree. The loyalty degree calculation unit 46 then outputs loyalty information indicating the degree of reliability or attachment of the user to the service or product from the relationship or degree between the specified factors.SELECTED DRAWING: Figure 4

Description

  The present invention relates to an information processing apparatus and a program.

  Patent Document 1 discloses an evaluation information extraction device that extracts and outputs evaluation information including target expressions, attribute expressions, and evaluation expressions from input text data.

  In Patent Document 2, a plurality of alternative models for obtaining loyalty from other history information such as browsing information other than the purchase history are generated for the existing model for obtaining loyalty from the purchase history of the product. A method for measuring the royalty of a non-user is disclosed.

  In Patent Document 3, a target value is set for each of two or more indexes such as loyalty, objective evaluation, subjective evaluation, characteristic, image, recognition degree, etc., and data indicating the result of branding activity is acquired. A brand value analysis apparatus that analyzes data by calculating a progress rate of a branding activity with respect to a target value based on data acquired in the above is disclosed.

  Patent Document 4 discloses a news material extraction device that automatically extracts social media information that can be a news material by supervised machine learning from a large number of social media information.

  In Patent Document 5, behavior attribute information is generated from access status to media, and media royalty information, which is information indicating the distribution of visitors repeatedly accessing based on the behavior attribute information, is generated. An information generation apparatus is disclosed that generates interest / preference information based on the characteristics of an article obtained from the analysis result of the above, and provides media loyalty information and interest / preference information as audience information.

JP 2008-140359 A JP 2012-247860 A JP 2017-045380 A JP 2017-201437 A JP-A-2018-005360

  The objective of this invention is providing the information processing apparatus and program which can grasp | ascertain possibility that a customer uses continuously with respect to a certain service or goods.

[Information processing device]
The present invention according to claim 1 is an acquisition means for acquiring customer transmission information for a certain service or product;
Classification means for classifying the transmission information acquired by the acquisition means for each type of factor representing the customer's feelings for the service or product;
A specifying means for specifying a relationship or a degree between a plurality of types of factors classified by the classification means;
And an output unit that outputs information indicating the degree of reliability or attachment of the customer to the service or product from the relationship or degree between the factors specified by the specifying unit.

  The present invention according to claim 2 relates to the switching of the plurality of factors to attitude factors related to customer attitudes toward products or services, behavioral factors related to customer purchase or usage behaviors regarding products or services, and other products or services. The information processing apparatus according to claim 1, comprising at least two types of switching factors.

  The present invention according to claim 3 is characterized in that the attitude factor is a factor related to a customer's attachment or dislike about a product or service, a factor related to an emotional expression expressing a customer's emotion, and the product or service to another customer. The information processing apparatus according to claim 2, wherein the information processing apparatus includes at least one of a factor regarding a recommendation intention which is a recommendation intention and a factor regarding a purchase intention which is a future purchase or use intention for the product or service.

  The invention according to claim 4 is the information according to any one of claims 1 to 3, wherein the specifying unit specifies different relations or degrees between the plurality of types of factors by performing different weighting on the plurality of types of factors. It is a processing device.

  The present invention according to claim 5 further comprises factor specifying means for specifying a factor that the customer has become such a feeling for each of a plurality of types of factors classified by the classification means. 4. The information processing apparatus according to any one of 4.

The present invention according to claim 6, wherein the plurality of factors include at least an attitude factor relating to a customer's attitude toward goods or services, and an action factor relating to a customer's purchase or usage behavior regarding goods or services,
The information processing apparatus according to claim 1, wherein the output unit outputs information indicating a degree of reliability or attachment of a customer to the service or the product from a combination of a relationship or a degree of the attitude factor and the behavior factor.

  In the present invention according to claim 7, the output means outputs that the degree of reliability or attachment of the customer is true when both the attitude factor and the behavior factor are high, and the behavior factor When the degree is high but the attitude factor is low, it outputs that the customer's reliability or attachment is a fake, and when the attitude factor is high but the behavior factor is low, The information processing apparatus according to claim 6, wherein the information indicating that the degree of reliability or the degree of attachment is potential is output.

[program]
The present invention according to claim 8 is an acquisition step of acquiring customer transmission information for a certain service or product;
Classification step of classifying the transmission information acquired in the acquisition step for each type of factor representing the customer's feelings for the service or the product;
A specifying step for specifying a relationship or degree between a plurality of types of factors classified in the classification step;
It is a program for causing a computer to execute an output step of outputting information indicating the degree of reliability or attachment of a customer to the service or product from the relationship or degree between factors specified in the specifying step.

  According to the first aspect of the present invention, it is possible to provide an information processing apparatus that can grasp the possibility that a customer continuously uses a certain service or product.

  According to the second aspect of the present invention, it is possible to provide an information processing apparatus that can grasp the possibility that a customer continuously uses a certain service or product.

  According to the third aspect of the present invention, it is possible to provide an information processing apparatus that can grasp the possibility that a customer continuously uses a certain service or product.

  According to the fourth aspect of the present invention, more accurate information is obtained as compared with the case where all the factors are treated uniformly to output information indicating the degree of reliability or attachment of the customer to the service or product. It is possible to provide an information processing apparatus that can

  According to the fifth aspect of the present invention, it is possible to provide an information processing apparatus capable of grasping a factor in which a feeling about a customer's service or product is formed.

  According to the sixth aspect of the present invention, it is possible to provide an information processing apparatus capable of grasping the type of customer reliability or attachment degree for a service or a product.

  According to the seventh aspect of the present invention, it is possible to provide an information processing apparatus capable of grasping the type of customer's reliability or degree of attachment to a service or product.

  According to the eighth aspect of the present invention, it is possible to provide a program that can grasp the possibility that a customer continuously uses a certain service or product.

It is a figure which shows the system configuration | structure of the royalty analysis system of one Embodiment of this invention. It is a block diagram which shows the hardware constitutions of the server apparatus 10 in one Embodiment of this invention. It is a block diagram which shows the function structure of the server apparatus 10 in one Embodiment of this invention. FIG. 4 is a diagram for explaining the functional configuration of a royalty measurement unit 32 shown in FIG. 3 in FIG. 4. It is a figure which shows the example of classification | category of the kind of factor showing the user's feelings with respect to a service or goods. It is a figure for demonstrating an example of the royalty information calculated by the royalty degree calculation part 46. FIG. It is a figure for demonstrating an example (transmission information example 1) of the transmission information acquired by the transmission information acquisition part 31. FIG. It is a figure for demonstrating the other example (transmission information example 2) of the transmission information acquired by the transmission information acquisition part 31. FIG. It is a figure for demonstrating the example of classification | category of the factor specified by the factor specific | specification part 33. FIG. It is a figure for demonstrating the evaluation example of the loyalty information by the combination of a behavior factor degree and an attitude factor degree.

  Next, embodiments of the present invention will be described in detail with reference to the drawings.

  FIG. 1 is a diagram showing a system configuration of a royalty analysis system including a server device 10 according to an embodiment of the present invention.

  As shown in FIG. 1, the royalty analysis system according to an embodiment of the present invention includes a server device 10, a server device 20, and a plurality of terminal devices 40 that are information processing devices connected via a network 30 such as the Internet. The Here, the server device 20 is, for example, an SNS (Social Networking Service) such as Twitter (registered trademark), blog, various electronic bulletin boards, and information posting sites that write reviews of various products and services. Network service), and provides SNS services to many users via the network 30.

  In addition, although this embodiment demonstrates using the case where the server apparatus 20 is operating the SNS site, it is operated by the company which provides various goods and services, and the user of the goods and services is provided. The present invention can be similarly applied even when a site where a customer can write comments and evaluations is operated.

  In such an SNS site, an information posting site, and the like, a lot of transmission information, which is information related to the product or service transmitted by the user as a customer, is written.

  And the server apparatus 10 which is a royalty analysis apparatus analyzes the loyalty (or loyalty) of the customer with respect to goods and a service by analyzing the various transmission information written in such an SNS site, an information contribution site, etc. Is going.

  Here, loyalty means the degree of trust or attachment of a user (customer) to the product or service. That is, for example, if loyalty is high for a product (high reliability or attachment), the product may be recommended to others, or a user with high loyalty can purchase the product continuously It means that the nature is high. In addition, when it is expressed as brand loyalty, it means the degree of trust and attachment that a user has to a product or service of a specific company brand.

  Such loyalty is used as an index for predicting whether or not the user will continue to purchase or use the product or service and whether or not the user recommends the product or service to others. For example, if the loyalty of a certain product or service is high (high reliability or attachment), the product or service may continue to be purchased or used in the future, or the product or service may be recommended to others. It can be judged that the nature is high. Conversely, if loyalty is low for a product or service (high reliability or attachment), the product or service may continue to be purchased or used in the future, or the product or service is recommended to others. In other words, it is determined that a user with low royalty is highly likely to use or purchase another product or service in the future.

  In addition, the terminal device 40 is a personal computer for each user who purchases or uses a product or service to write transmission information such as evaluations and comments on the product or service on an SNS site provided by the server device 20. (Hereinafter abbreviated as PC). Although FIG. 1 shows the case where the terminal device 40 is a personal computer, the terminal device 40 may be another device such as a mobile phone, a smartphone, or a tablet terminal device.

  Next, FIG. 2 shows a hardware configuration of the server device 10 that functions as a royalty analysis device in the royalty analysis system of the present embodiment.

  As shown in FIG. 2, the server device 10 communicates with a CPU 11, a memory 12, a storage device 13 such as a hard disk drive (HDD), and data transmission and reception with an external device via a network 30. It has an interface (IF) 14, a touch panel or liquid crystal display, and a user interface (UI) device 15 including a keyboard. These components are connected to each other via a control bus 16.

  The CPU 11 executes predetermined processing based on a control program stored in the memory 12 or the storage device 13 to control the operation of the server device 10. In the present embodiment, the CPU 11 has been described as reading and executing a control program stored in the memory 12 or the storage device 13, but the program is stored in a storage medium such as a CD-ROM and stored in the CPU 11. It is also possible to provide.

  FIG. 3 is a block diagram showing a functional configuration of the server device 10 realized by executing the control program.

  As shown in FIG. 3, the server device 10 according to the present embodiment includes a transmission information acquisition unit 31, a royalty measurement unit 32, a factor identification unit 33, a royalty information storage unit 34, a factor information storage unit 35, A statistical processing unit 36 is provided.

  The transmission information acquisition unit 31 acquires transmission information of a user (customer) for a certain service or product. Specifically, the transmission information acquisition unit 31 is a specific service or a service for which loyalty is to be investigated from the posting information posted on the SNS and various information posting sites, etc. Get outgoing information about the product. Note that the transmission information acquired by the transmission information acquisition unit 32 includes retweet information that quotes the posted content (tweet) of another person on Twitter (registered trademark).

  The royalty measurement unit 32 measures the royalty, which is the user's reliability or attachment level, for the service or product for which the transmission information is acquired based on the transmission information acquired by the transmission information acquisition unit 31, and this reliability or attachment level Royalty information is generated as information indicating the.

  The royalty information storage unit 34 stores the royalty information generated by the royalty measurement unit 32.

  The factor identifying unit 33 identifies the factor that caused the customer to have such a feeling for various expressions included in the transmission information acquired by the transmission information acquiring unit 31, and outputs the identification result as factor information. The specific operation of the factor specifying unit 33 will be described later.

  The factor information storage unit 35 stores the factor information output by the factor specifying unit 33.

  The statistical processing unit 36 performs statistical processing using the royalty information stored in the royalty information storage unit 34 and the factor information stored in the factor information storage unit 35.

  Further, the statistical processing unit 36 aggregates loyalty information for each product or service or loyalty information for each user, and the trend of the user's loyalty with respect to the product or service and why such loyalty is obtained. Analyzing factors such as

  Further, the statistical processing unit 36 adjusts the royalty standard in the royalty measurement unit 32 using the royalty information and the factor information.

  Specifically, the statistical processing unit 36 calculates an average value of the degree of royalty (degree of reliability or degree of attachment) for a plurality of products or services of a certain user, and the average value is higher than that of other users. For the user, adjustments are made to increase the royalty criteria. Conversely, the statistical processing unit 36 performs an adjustment to lower the loyalty determination criterion for a user whose average value of the degree of royalty for a plurality of products or services of a certain user is lower than that of other users.

  In other words, the statistical processing unit 36 has a higher frequency of sending good comments and evaluations about products or services for users whose evaluation criteria for products or services are looser than other users, so that the criteria for determining royalties are slightly higher. Adjust to. In addition, the statistical processing unit 36 has a high frequency of sending bad comments or evaluations about products or services for users whose evaluation criteria for products or services are stricter than other users. adjust.

  Next, the functional configuration of the royalty measurement unit 32 shown in FIG. 3 will be described with reference to FIG.

  As shown in FIG. 4, the royalty measurement unit 32 includes a royalty expression extraction unit 41, an expression classification unit 42, an attitude factor degree calculation unit 43, a behavior factor degree calculation unit 44, a switching factor degree calculation unit 45, The loyalty degree calculation unit 46 is configured.

  The loyalty expression extraction unit 41 extracts expressions related to the degree of reliability or love of the user's service or product from the transmission information acquired by the transmission information acquisition unit 31. Specifically, the loyalty expression extraction unit 41 uses lexical features (word strings, morpheme strings, etc.), structural features (partial syntax trees, etc.), semantic features (meaning of words) in outgoing information. Then, machine learning is performed by a supervised learning method using evaluation expressions and the like, and an extraction model is generated. Then, the loyalty expression extraction unit 41 uses the generated extraction model to extract an expression (loyalty expression) related to the degree of reliability or affection for the service or product from the user's transmission information.

  The loyalty expression extraction unit 41 may generate an extraction model by performing machine learning using an unsupervised learning method. Further, the loyalty expression extraction unit 41 does not use the extraction model, but uses, for example, information that is set in advance as to what expression is the expression relating to the reliability or the affection degree for the service or the product. It is also possible to extract loyalty expressions from the information.

  The expression classification unit 42 classifies the transmission information acquired by the transmission information acquisition unit 31 for each type of factor representing the user's feelings about the service or the product. FIG. 5 shows an example of classification of the factor types.

  In the example shown in FIG. 5, as major categories of factors, an attitude factor related to a customer's attitude toward a product or service, a behavioral factor related to a customer's purchase or use behavior for a product or service, and switching related to switching to another product or service Three types of factor classification of factors (switching factors) are shown.

  And as a subcategory of attitude factors, it is a factor related to the customer's attachment or dislike about the product or service, a factor related to the emotional expression expressing the customer's emotion, and the intention to recommend the product or service to other customers Four types of factor classifications are shown: a factor relating to a recommendation intention and a factor relating to a purchase intention which is a future purchase or use intention for the product or service.

  Note that royalty analysis can be performed by including at least one of the four types of factor classifications in the attitude factor.

  In addition, as a small classification of behavior factors, usage factors relating to the current usage / usage status, frequency, etc. of goods or services are shown.

  Furthermore, as a subcategory of switching factors, there are switching barrier factors related to the difficulty or intention to switch to other products or services, and competing product factors related to other products or services such as comparison with other competing products or services. It is shown.

  In the present embodiment, three major factor classifications, that is, an attitude factor, a behavior factor, and a switching factor, will be described as the major factor categories. However, if at least two of these three factor categories are used. It is possible to do loyalty analysis.

  The expression classification unit 42 generates a classification model for determining what expression is classified into which factor by machine learning using a supervised learning method. Then, the expression classification unit 42 determines to which factor the various expressions extracted by the loyalty expression extraction unit 41 should be classified using the generated classification model. Details of the expression factor classification by the expression classification unit 42 will be described later.

  The attitude factor degree calculation unit 43 calculates an attitude factor degree based on the expression calculated by the expression classification unit 42 as an attitude factor. This attitude factor degree means a user's reliability or attachment degree to a product or service based on an expression classified as an attitude factor.

  The behavior factor degree calculation unit 44 calculates the behavior factor degree based on the expression calculated by the expression classification unit 42 as a behavior factor. This behavior factor degree means a user's reliability or attachment degree to a product or service based on an expression classified as a behavior factor.

  The switching factor degree calculation unit 45 calculates the switching factor degree based on the expression calculated by the expression classification unit 42 as a switching factor. This switching factor degree means a user's reliability or attachment degree to goods or services based on expressions classified as switching factors.

  The royalty degree calculation unit 46 uses the attitude factor degree, the behavior factor degree, and the switching factor degree calculated by the attitude factor degree calculation unit 43, the behavior factor degree calculation unit 44, and the switching factor degree calculation unit 45, respectively. The loyalty information which shows the user's total reliability or attachment degree with respect to goods is output.

  Specifically, the loyalty degree calculation unit 46 specifies the relationship or degree between a plurality of types of factors classified by the expression classification unit 42 from the attitude factor degree, the behavior factor degree, and the switching factor degree. Then, the loyalty degree calculation unit 46 outputs loyalty information indicating the degree of reliability or attachment of the user to the service or the product from the relationship or degree between the specified factors.

  The loyalty level calculation unit 46 does not calculate loyalty information by adding the three types of factor degrees of attitude factor degree, behavior factor degree, and switching factor degree with the same weighting, but differs for the three types of factors. Weighting may be performed to specify the relationship or degree between the three types of factors.

  Next, an example of loyalty information calculated by the loyalty degree calculation unit 46 will be described with reference to FIG.

  The example of loyalty information shown in FIG. 6 shows a case where the degree of loyalty (the degree of reliability or attachment of a user to a service or a product) is represented by numerical information called a loyalty level. For example, the royalty level is set in five stages from +2 to -2, and the royalty level calculation unit 46 outputs a value of +2 when the royalty of the transmission information to be determined is very high.

  Next, a specific example in which the royalty measurement unit 32 measures the royalty from the transmission information acquired by the transmission information acquisition unit 31 will be described with reference to FIGS.

  For example, when measuring royalty based on the transmission information example 1 as shown in FIG. 7, the royalty measurement unit 21 uses the royalty expression extraction unit 41 to extract the attitude factor, behavior factor, and switching factor from the transmission information example 1. Extract words, phrases, and sentences.

  Then, the expression classification unit 42 determines whether each extracted factor belongs to an attitude factor, a behavior factor, or a switching factor, and further, to which of the major categories of each factor the attribute is attributed. And classify.

  Then, the attitude factor degree calculation unit 43, the behavior factor degree calculation unit 44, and the switching factor degree calculation unit 45 calculate each factor degree for each factor.

  As an example, the attitude factor degree calculation unit 43, the behavior factor degree calculation unit 44, and the switching factor degree calculation unit 45 add 1 point for a positive expression and subtract 1 point for a negative expression, Factors are calculated by summing points for each type of factor.

  For example, the attitude factor degree calculation unit 43 has a point of “−1” for a negative attitude factor of an emotional expression “the staff manners and the fee system are worse every year” extracted from the transmission information. Is added. And the attitude factor degree calculation part 43 calculates the total value added with the point added to the other expression of an attitude factor as an attitude factor degree.

  Similarly, the behavior factor degree calculation unit 44 and the switching factor degree calculation unit 45 similarly calculate the behavior factor degree and the switching factor degree by adding the points added to the expressions classified as the behavior factor and the switching factor. To do.

  Each factor degree calculated by the attitude factor degree calculating unit 43, the behavior factor degree calculating unit 44, and the switching factor degree calculating unit 45 is provided with an upper limit and a lower limit, and even if the number of classified expressions is large, For example, the range of −3 to +3 may be limited.

  In the transmission information example 1 shown in FIG. 7, since many negative expressions are included in each of the attitude factor, the behavior factor, and the switching factor, the royalty level calculated by the royalty level calculation unit 46 is, for example, −2. Thus, the value indicates that loyalty is low.

  Moreover, the other example of the transmission information acquired by the transmission information acquisition part 31 is shown in FIG.

  Also for the transmission information example 2 shown in FIG. 8, the royalty measurement unit 21 uses the royalty expression extraction unit 41 to express words, phrases, and sentences from the transmission information example 1 as attitude factors, behavior factors, and switching factors. To extract. However, in this transmission information example 2, there is no expression classified as a switching factor.

  Then, the attitude factor degree calculation unit 43 and the behavior factor degree calculation unit 44 determine whether each of the classified expressions is a positive expression or a negative expression, give points, and give the given points. The factor is calculated by summing.

  For example, in the example shown in FIG. 8, the attitude factor calculation unit 43 extracts the document “accent has become natural in half a year even though it was initially a simple katakana English conversation” extracted from the transmission information. In addition, it is shown that 1 point is given to each phrase “It seems to be fun to attend”.

  Then, the attitude factor degree calculation unit 43 and the behavior factor degree calculation unit 44 calculate the total value of the attitude factor and additional points for each behavior factor as the attitude factor degree and the behavior factor degree. Then, the loyalty degree calculation unit 46 calculates the final loyalty degree based on the calculated attitude factor degree and behavior factor degree.

  In addition, in the transmission information example 2 shown in FIG. 8, since many positive expressions are included in each of the attitude factor and the behavior factor, the royalty level calculated by the royalty level calculation unit 46 is, for example, +2, such as +2. The value indicates a high value.

  Next, the operation of the factor specifying unit 33 shown in FIG. 3 will be specifically described. The factor specifying unit 33 specifies the factor that caused the customer to have such a feeling for each of a plurality of types of factors extracted by the royalty expression extracting unit 41 of the royalty measuring unit 32 and classified by the expression classifying unit 42. .

  Specifically, the factor specifying unit 33 determines which factor the user has become such a feeling for each factor extracted from the transmission information based on the factor classification example as shown in FIG. Identify whether it belongs to a classification.

  For example, when a factor “not used anymore” is extracted from the transmission information, the factor specifying unit 33 determines whether the factor of the factor is related to “function” or “quality”. ", Or about other contents such as" convenience "and" staff ".

  As a result, in the statistical processing unit 36, for example, a negative expression for a certain product or service is ranked for each factor specified by the factor specifying unit 33, thereby grasping items that have lost the reputation of the product or service. Will be able to.

  In the embodiment described above, the loyalty information output by the loyalty measurement unit 32 is a loyalty level value indicating the degree of loyalty, but the loyalty information is not limited to this.

  Another example of such loyalty information evaluation method will be described with reference to FIG.

  In the evaluation example shown in FIG. 10, the loyalty degree calculation unit 46 evaluates loyalty by a combination of two kinds of factor degrees, ie, a behavior factor degree and an attitude factor degree.

  That is, the plurality of factors must include at least an attitude factor related to the customer's attitude toward the product or service and an action factor related to the purchase or usage behavior of the customer regarding the product or service.

  Then, the loyalty degree calculation unit 46 outputs information indicating the degree of reliability or attachment of the customer to the service or the product from the relationship between the attitude factor and the behavior factor or a combination of the degrees.

  Specifically, the loyalty level calculation unit 46 outputs that the loyalty of the user is a true loyalty indicating that the loyalty is true when both the attitude factor and the behavior factor are high. Further, when the degree of behavior factor is high but the degree of attitude factor is low, the loyalty level calculation unit 46 outputs that the loyalty is a sham loyalty indicating that the user's loyalty is a sham. Further, when the degree of attitude factor is high but the degree of behavior factor is low, the loyalty degree calculation unit 46 outputs that the loyalty is a potential loyalty indicating that the user's loyalty is a potential one. When both the attitude factor and the behavior factor are low, the loyalty degree calculation unit 46 outputs that there is no royalty indicating that the loyalty of the user is low.

  By evaluating loyalty based on such a combination of behavior factor and attitude factor, for example, if advertising information is distributed, effective acquisition of new customers will be targeted for potential loyalty users. Can be done. Further, for example, if the separation from goods or services is to be prevented, it is possible to effectively prevent the current user from leaving by targeting a user with a fake royalty.

10 Server device 11 CPU
12 Memory 13 Storage Device 14 Communication Interface (IF)
DESCRIPTION OF SYMBOLS 15 User interface (UI) apparatus 16 Control bus 20 Server apparatus 30 Network 31 Transmission information acquisition part 32 Loyalty measurement part 33 Factor specification part 34 Royalty information storage part 35 Factor information storage part 36 Statistical processing part 40 Terminal apparatus 41 Royalty expression extraction part 42 Expression classification unit 43 Attitude factor calculation unit 44 Behavior factor calculation unit 45 Switching factor calculation unit 46 Loyalty calculation unit

Claims (8)

An acquisition means for acquiring customer transmission information for a service or product;
Classification means for classifying the transmission information acquired by the acquisition means for each type of factor representing the customer's feelings for the service or product;
A specifying means for specifying a relationship or a degree between a plurality of types of factors classified by the classification means;
Output means for outputting information indicating the degree of reliability or attachment of the customer to the service or product from the relationship or degree between the factors specified by the specifying means;
An information processing apparatus comprising:   The plurality of factors are at least two or more of an attitude factor relating to a customer's attitude to a product or service, a behavior factor relating to a customer's purchase or use behavior for the product or service, and a switching factor relating to switching to another product or service. The information processing apparatus according to claim 1 including:   The attitude factor is a factor relating to the customer's attachment or dislike about the product or service, a factor relating to the emotional expression expressing the customer's emotion, and a factor relating to the recommendation intention which is an intention to recommend the product or service to other customers. The information processing apparatus according to claim 2, further comprising: at least one of factors relating to purchase intention, which is future purchase or use intention for the product or service.   The information processing apparatus according to any one of claims 1 to 3, wherein the specifying unit specifies different relationships or degrees between the plurality of types of factors by performing different weighting on the plurality of types of factors.   The information processing apparatus according to any one of claims 1 to 4, further comprising factor specifying means for specifying a factor that causes a customer to have such a feeling for each of a plurality of types of factors classified by the classification means. . The plurality of factors include at least an attitude factor related to a customer's attitude toward a product or service, and an action factor related to a customer's purchase or usage behavior regarding the product or service,
The information processing apparatus according to claim 1, wherein the output unit outputs information indicating a customer's reliability or attachment degree to the service or the product from a combination of a relationship or a degree of the attitude factor and the behavior factor.   When both the attitude factor and the behavior factor are high, the output means outputs that the customer's reliability or attachment is true, and the behavior factor is high but the attitude factor is Is low, the customer's reliability or attachment is output, and if the attitude factor is high but the behavior factor is low, the customer's reliability or attachment is potential The information processing apparatus according to claim 6, which outputs a message indicating that it is a product. An acquisition step of acquiring customer transmission information for a service or product;
Classification step of classifying the transmission information acquired in the acquisition step for each type of factor representing the customer's feelings for the service or the product;
A specifying step for specifying a relationship or degree between a plurality of types of factors classified in the classification step;
From the relationship or degree of the factors specified in the specifying step, an output step for outputting information indicating the degree of reliability or attachment of the customer to the service or product;
A program that causes a computer to execute.