JP6945019B2 - Customer estimation device and customer estimation method - Google Patents

Description

The present invention relates to a customer estimation device and a customer estimation method capable of accurately estimating a target customer and verifying the reliability of the result.

Conventionally, a system has been known that uses a web access log to deliver or propose an advertisement for a product or service to a new customer.

For example, in the following targeting advertisement providing system for members of Patent Document 1, the members of the member-only site A operated by the member-only site operating company are members-only from a user terminal such as a computer or a smartphone via the Internet. Access the site main server. The member can log in to the member-only site A by entering the member ID and password from the user terminal and browse the information providing page for the member.

Then, the user terminal passes the cookie ID of the member-only site non-member compatible server separated from the member-only site A to the advertisement distribution server of the advertisement distribution company, so that the history in the member-only site A is leaked to the outside. Providing targeted advertisements that match the attributes of members while preventing this from happening.

In addition, when the member accesses the website B operated by the website operating company from the user terminal via the Internet, the advertisement is delivered to the ad space of the website B that has been entrusted with the advertisement distribution. (Paragraphs 0017 to 0027, FIG. 1).

Patent No. 5843983

However, in the system of Patent Document 1, since the advertisement is delivered based on the information acquired from the cookie ID, there is a problem that only advertisements for highly related products and services such as related products of the products purchased by the user can be performed. was there. In other words, it was not possible to make proposals for solving matters that users have recently become interested in or potential problems. In order to make such a proposal, it was premised on the acquisition of a concatenated identifier that identifies an individual.

The present invention has been made in view of the above problems, and an object of the present invention is to provide a customer estimation device capable of accurately estimating a target customer without acquiring a concatenated identifier.

It should be noted that this case is an application related to a customer estimation device (Japanese Patent Application No. 2019-162873) for which an application has been filed, and the purpose is to explain the practicality and expandability of the device.

The customer estimation device of the present invention has acquired an indexed item acquisition means for acquiring an indexed item to be indexed, a customer-owned item acquisition means for acquiring a customer-owned item owned by a customer, and the customer-owned item acquisition means. Obtained from the customer-owned items and the survey conducted on a specific or unspecified person or the history information of that person, and reflects at least one of the attributes, values, preferences, consumption behavior, and behavior pattern of that person. According to the item corresponding means for associating the source data item possessed by the source data, the machine learning model in which the source data item corresponding to the customer-owned item is machine-learned as input data, and the content of the indexed item. Then, the classification item creation means for classifying the customer-owned items, which are elements necessary for estimating the target customer who will be in demand in the future using the machine learning model, and creating the classification items, and the classification item creation means The index is obtained by referring to a score-giving means for assigning a score to the created classification item based on a predetermined rule using the machine learning model, and an information database having information related to the indexed item. wherein the occurrence rate estimating means for estimating the incidence of target customers to estimate the number of the target customer from said occurrence rate the occurrence rate estimating means has estimated the total number of customers, the reply to the indexing item for reduction item It is characterized by including a customer determination means for determining the target customer by selecting in order from the one with the highest score given to each item of the data group having the above.

In the customer estimation device of the present invention, the classification item creating means creates classification items according to the contents of indexing items related to products to be sold, services to be provided, and the like. Further, the scoring means assigns a score to each classification item based on a predetermined rule. As a result, the present device can distinguish between the classification item having a high score of the indexing item and the classification item having a low score of the indexing item, and can estimate the candidate of the target customer.

Further, in the present invention, the appearance rate estimation means estimates the appearance rate of the target customer for the indexed item in advance by referring to the information database. Furthermore, the score giving means gives a score to the data group having the answer about the same index item, the customer judgment means estimates the number of target customers, and the one with the highest score is judged as the target customer according to the appearance rate. do. In this way, by acquiring the appearance rate in advance, it is possible to accurately estimate the actual target customer.

Even if the customer estimation device of the present invention includes a reliability verification means for verifying the reliability calculated from the degree of agreement obtained by comparing the answer with the result of the target customer determined by the customer determination means. good.

The reliability verification means compares the answer (for example, the questionnaire result) possessed by the data group with the result of the target customer determined by the customer determination means, and the more the two match, the higher the accuracy of the determination. In this way, the present device can verify the estimated reliability of the target customer.

Further, in the customer estimation device of the present invention, when the customer determination means exceeds the number of people corresponding to the appearance rate due to the existence of a plurality of the customers having the same score on the determination line score determined from the appearance rate. , The number of the target customers may be adjusted by arbitrarily extracting from the customers.

In this device, there may be a plurality of customers on the determination line score determined from the appearance rate, and it may not be possible to accurately select the target customer according to the appearance rate. In such a case, the customer determination means arbitrarily extracts several customers from the customers with the same score and adjusts (reduces) the number of target customers. As a result, the present device can select an appropriate number of target customers.

The customer estimation method of the present invention is a customer estimation method for estimating a target customer who is interested in a product or service by using a computer provided with a machine learning model, and the data acquisition unit of the computer is an index to be indexed. An indexed item acquisition step for acquiring an item, a customer-owned item acquisition step for the data acquisition unit to acquire a customer-owned item owned by the customer, and a data processing unit for the computer to acquire the customer-owned item in the customer-owned item acquisition step. Obtained from the customer-owned items and the survey conducted on a specific or unspecified person or the history information of that person, and at least one of the attributes, values, preferences, consumption behavior, and behavior pattern of that person An item correspondence step that associates the source data item of the reflected source data, and a machine learning model creation step that creates a machine learning model by machine learning the source data item corresponding to the customer-owned item as input data. , The data processing unit classifies the customer-owned items, which are necessary elements for estimating the target customer who will be in demand in the future, using the machine learning model according to the contents of the indexed items, and classifies the items. A classification item creation step for creating the above, a score assignment step in which the data processing unit assigns a score to the classification item created in the classification item creation step based on a predetermined rule using the machine learning model. wherein the data processing unit, by referring to the information database to hold information related to the indexing item, the appearance rate estimation step of estimating the incidence of the target customer for the indexing item, wherein the data processing unit the estimates the number of the target customer from the appearance rate estimating the incidence estimated in step and the total number of customers, high the scores given to each item of the data groups having a reply to the indexed item It is characterized by including a customer determination step of determining the target customer by selecting in order from the other.

In the customer estimation method of the present invention, a classification item is created according to the indexed item in the classification item creation step, and a score is given to each classification item in the score giving step. Thereby, this method can distinguish the classification item having a high score of the indexing item from the classification item having a low score of the indexing item, and can estimate the candidate of the target customer.

Further, in the appearance rate estimation step of this method, the appearance rate of the target customer for the indexed item is estimated in advance by referring to the information database. Furthermore, in the score giving step, a score is given to the data group having the answer about the same index item, and in the customer judgment step, the number of target customers is estimated, and the highest score is given according to the appearance rate. Determine the target customer. In this way, by acquiring the appearance rate in advance, it is possible to accurately estimate the actual target customer.

In the customer estimation method of the present invention, the reliability verification that the data processing unit verifies the reliability calculated from the degree of agreement obtained by comparing the answer with the result of the target customer determined in the customer determination step. It may have steps.

In the reliability verification step, the answer of the data group is compared with the result of the target customer determined in the customer determination step, and the more they match, the higher the accuracy of the determination. In this way, the estimated reliability of the target customer can be verified by this method.

Further, in the customer estimation method of the present invention, in the customer determination step, when a plurality of the customers having the same score are present on the determination line score determined from the appearance rate, the number of people corresponding to the appearance rate is exceeded. , The number of the target customers may be adjusted by arbitrarily extracting from the customers.

In this method, there may be a plurality of customers on the determination line score determined from the appearance rate, and it may not be possible to accurately select the target customer according to the appearance rate. In such a case, in the customer determination step, several customers with the same score are arbitrarily extracted to match (reduce) the number of target customers. As a result, the method can select an appropriate number of target customers.

The figure explaining the outline of the customer estimation apparatus which concerns on embodiment of this invention. The figure explaining the example of the source data item. The figure explaining the example of the classification item created by the customer estimation apparatus. The figure explaining the example of the customer list which added the judgment result. The figure explaining the example of the information database which holds the indexed item. A flowchart for confirming the reliability of a target customer using a customer estimation device.

Hereinafter, the details of the customer estimation device of the present invention will be described with reference to the drawings.

First, the outline of the customer estimation device 1 according to the embodiment of the present invention will be described with reference to FIG. The customer estimation device 1 of the present embodiment is a device terminal installed on a desktop PC, a notebook PC, a tablet terminal, or the like and owned by a service provider X (hereinafter referred to as company X).

In the following, the client company of company X, the service provider Y of the customer estimation device 1 (hereinafter referred to as company Y), distributes DM (direct mail) and advertises online for newly developed pharmaceutical products. To get started, let's take as an example the case where there is a demand to identify a person suffering from back pain.

Company Y has a customer DB (database) 20, and has at least "gender" and "age" as customer information in the customer list. However, it does not have direct information such as "whether or not you are suffering from back pain".

In such a case, the company Y (user) transmits to the company X an item to be indexed (indexed item) such as "presence or absence of back pain worries" and a customer-owned item of "gender" and "age". Here, the "gender" of the customer-owned item is male / female information, and the "age" is information such as teens / 20s / 30s / 40s / 50s / 60s, and "customer AAA is not necessarily". It does not have to be personal information such as ", female, 35 years old".

By inputting these items into the customer estimation device 1 by company X (person in charge), not only the prospective customers related to the indexed items from all the customers but also the possibility is low at present, but the demand will increase in the future. Target customers (potential customers or target customers) that are presumed to exist can be extracted.

The indexed item is acquired by the indexed item acquisition unit 2 of the customer estimation device 1 (corresponding to the “indexed item acquisition means” of the present invention). Further, the customer-owned items are acquired by the customer-owned item acquisition unit 3 of the customer estimation device 1 (corresponding to the "customer-owned item acquisition means" of the present invention).

When indexed items and customer-owned items are input to the customer estimation device 1, the item correspondence unit 4 (corresponding to the "item correspondence means" of the present invention) of the customer estimation device 1 can estimate the entire market by area random sampling. Matches items with single source data (hereinafter referred to as source data S) that employs. Further, the customer estimation device 1 classifies the customer-owned items and assigns a score to each classification item.

The customer-owned items are classified by the classification item creation unit 5 of the customer estimation device 1 (corresponding to the "classification item creation means" of the present invention). Further, the score is given by the score giving unit 6 of the customer estimation device 1 (corresponding to the "score giving means" of the present invention). The classification item creation unit 5 and the score giving unit 6 are machine learning models in which the source data S is machine-learned as input data. It can be said that this is the provision of information on indexed items rather than information on target customers.

Supplementing the source data S, this is information that comprehensively investigates the consciousness, attributes, product involvement, media contact, etc. of specific or unspecified consumers by questionnaire. That is, the source data S is information that strongly reflects the behavior patterns of consumers, which captures consumers from the three viewpoints of "people," "things," and "media."

The above-mentioned "consciousness" means the values, tastes, consumption behaviors, etc. of consumers. In addition, "attribute" is, to be exact, a demographic attribute, such as gender, age (age), area of residence, industry and annual income if working, unmarried or married, how many children there are, etc. Means the socio-economic characteristic data of.

In addition, the above-mentioned "product involvement" means data such as brands of clothes and electrical appliances, hobbies and special skills, and "media contact" means whether the medium used for acquiring information is the Internet or newspapers. Or it means data such as a magazine.

Source data S surveyed about 10,000 people extracted from 7 areas, which are the main markets in Japan, in the Tokyo area (within 50 km of Tokyo), Kansai area, Nagoya area, northern Kyushu area, Sapporo area, Sendai area, and Hiroshima area. It was obtained by renting and implementing a dedicated tablet. The source data S is not limited to the response to the questionnaire, and may include viewing log data (history information) such as an audience rating.

Further, the customer estimation device 1 includes an appearance rate estimation unit 7 (corresponding to the "appearance rate estimation means" of the present invention) that estimates the appearance rate of the target customer (customer suffering from back pain) of the indexed item. .. The details will be described later, but the appearance rate estimation unit 7 uses the appearance rate in the appropriate information DB (database) 30 that holds the indexed items this time (for example, answers "with back pain" in the questionnaire). The appearance rate of the target customer in the customer DB 20 is estimated.

Further, the customer estimation device 1 includes a customer determination unit 8 (corresponding to the "customer determination means" of the present invention) that determines a target customer. The customer determination unit 8 determines the optimum number of target customers from the data group of the information DB 30 by using the appearance rate of the target customers estimated by the appearance rate estimation unit 7 as a threshold value.

Further, the customer estimation device 1 includes a reliability verification unit 9 (corresponding to the “reliability verification means” of the present invention) that verifies the reliability of the execution result. Specifically, the reliability verification unit 9 calculates the reliability of the execution result by using the target customer information (answers made by the target customer) acquired in advance by the appearance rate estimation unit 7.

Next, FIG. 2 shows an example of a data item (source data item) of the source data S owned by the company X.

As described above, the source data S consists of a set of questionnaire results and includes many elements, but has a data item of "back pain" as an item related to this index item (whether or not there is a problem of back pain). ..

In this example, it means that the individual of "Sample No. 1 and 2" checked "with back pain" in the questionnaire, and the individual of "Sample No. 3" did not check "with back pain". do. The "sample number" is a random number and does not identify an individual.

According to the present invention, the reliability of normal execution of the customer estimation device 1 can be estimated if there is an appropriate database having indexed items. However, there are cases where a convenient database is not available. In such a case, the source data S may be used as a substitute to estimate the appearance rate of the target customer.

Next, FIG. 3 shows an example of a classification item (list 15) created by the customer estimation device 1 from the customer-owned items. The customer estimation device 1 classifies the customer-owned items with respect to the indexed items by the above-mentioned classification item creation unit 5 (rules possessed by the machine learning model).

By this work, the first category item "Gender [male] & age [50s and over]", the second category item "Gender [male] & age [10-40s]", and the category item 3 The third item, "Gender [female] (regardless of age)," is created.

Further, the score giving unit 6 of the customer estimation device 1 assigns a score to each classification item. This score is calculated from the numerical value of each classification item with respect to the reference value when the reference value for the entire customer is calculated and this is set to 1 (relative evaluation). For example, the first classification item is a group showing a numerical value 2.5 times larger than the reference value, and basically, it can be estimated to be a group with a high score that correlates with the indexed item. For this reason, it is most efficient to deliver DMs of low back pain medicines and treatment methods to the first customer (provisional target customer) in the classification item.

Next, FIG. 4 shows an example of the customer list 16 in which the customer estimation device 1 is executed and the items of “score” and “judgment” are added.

The score giving unit 6 of the customer estimation device 1 inputs a score (backache score) to each classification item for 10 customers (customer IDs: AAA to JJJ). Then, the customer determination unit 8 assigns "○" or "x" to the "determination (low back pain determination)" column for the customer who estimates that the degree of low back pain is high (for example, the threshold value is 2.0). do. This is the determination result of the normal execution of the customer estimation device 1.

On the other hand, the score giving unit 6 of the customer estimation device 1 executes the score giving to the data group of the information DB 30. FIG. 5 shows an example of an information list 17 in which the information DB 30 (personal identifier: ah-kokoko) is sorted in descending order of score.

In the information list 17, the source data ID (ah) with a "score" of 2.5 is listed at the top, and the source data ID with a "score" of 1.3 and the source data with a "score" of 0.5 are listed below. Followed by ID (the "age" and "gender" columns are omitted).

Further, since the information list 17 holds the information of the target customers of the indexed items (column of "backache"), the appearance rate estimation unit 7 acquires the appearance rate for all the customers. Here, the estimation is based on the assumption that the appearance rate of the information DB 30 and the appearance rate of the customer DB 20 are equal, but this is only an example. Other methods that can estimate the appearance rate in the customer DB 20 may be used. By this work, it is possible to estimate in advance what percentage of all customers can be the target customers. It is assumed that the appearance rate this time was 40%.

In the information list 17, the appearance rate of the target customer of the estimated customer DB 20 was 40%, so the top 40% of the score is used as the determination line (threshold value) to determine the target customer. Therefore, the top four scores are selected.

Since there are four personal identifiers with a score of "1.3" on the determination line this time, the customer determination unit 8 randomly selects three from these four and assigns them "○". As a result, four personal identifiers "Ah", "Ii", "Uuu", and "Yes" are selected, and "○" is given to the "Judgment" column.

After that, the reliability verification unit 9 confirms the coincidence and disagreement between the column of "back pain" in the information list 17 and the column of "determination (decision of back pain)" which is the final result. According to the information list 17, the results are the same except for the source data ID "Uu" among the four top scores, so the reliability of the determination by the customer estimation device 1 this time is 75%.

If the four personal identifiers "Ah", "Ii", "Uu", and "Oh" are selected, the results of the personal identifier "Oh" will be inconsistent, and the reliability of the judgment will decrease. become. In this example, since the number of personal identifiers is as small as 10, the reliability is affected, but in reality, there are many cases where the number of customers is 100 or more, and the effect of the reliability due to the random number adjustment is small.

In this way, it is preferable to extract the target customers and deliver the DM of the new drug to the customers. Further, the customer estimation device 1 is advantageous in that the customer profile can be accurately estimated without obtaining personal information, cookie information, etc., which are difficult to handle.

Finally, with reference to FIG. 6, a general flowchart until the client company (for example, the above-mentioned company Y) acquires the information of the target customer by using the customer estimation device 1 will be described.

First, the user of the client company selects the indexing item (STEP 10). The indexing item may be selected from the source data by the person in charge, and it is not limited to the provision of products and services performed by the client company, but a wide range of themes such as whether or not there are people with the same hobby. It may be. After that, the user transmits the indexed item and the customer-owned item to the service providing company (for example, the above-mentioned company X).

Next, at the service provider, the person in charge inputs the indexed item and the customer-owned item into the customer estimation device 1 (STEP 20). After that, the customer estimation device 1 adjusts the customer-owned item and the source data item (STEP30). Here, the items owned by the customer include "gender", "age (age)", "final educational background", "student / worker", "area of residence", etc., and the larger the number of items, the more the source. The possibility of matching (corresponding) with the data item is increased, and the accuracy of estimation is improved.

In STEP 30, for example, when both items do not completely correspond to each other, but there are similar items, the work of aligning the meanings is performed. For example, if the source data item has an item related to "favorite program genre" and the customer-owned item has an item "view the genre three times or more in a month (history information)", both items are treated as synonymous. ..

When adjusting both items, it may be adjusted to match the item of the higher concept (for example, the coarser grain size). For example, when the source data item has an item of "detached house (new construction) or detached house (used)" and the customer-owned item has an item of "detached house", the latter can be said to be a higher-level conceptual item. Therefore, the latter item is adopted. By adjusting the items as described above, the source data items and the customer-owned items are well linked, so it is possible to create classification items in relation to the adjusted items and the indexed items. ..

Next, at the service provider, the customer estimation device 1 creates a classification rule by machine learning (STEP 40). As a result, a classification item for the indexed item is created. Subsequently, the customer estimation device 1 aggregates the scores based on the rule (STEP 50). Specifically, a score is given to each classification item, and a list 15 (see FIG. 3) is created.

After that, at the service provider, the customer estimation device 1 estimates the appearance rate of the target customer (STEP 60). For example, the appearance rate of the target customer for the indexed item is examined by using an appropriate database such as the information DB 30 or the source data S owned by the service provider.

Next, a score is given for each customer ID in the customer list (STEP 70). Specifically, using the list 15, a score is input for each customer by a program and indexed (see FIG. 4).

Next, "○" or "×" is assigned to each customer ID based on the score (STEP80). Here, "○" means that it is presumed to be a target customer, and for example, when the score exceeds (or exceeds the threshold value) the threshold value derived from the appearance rate, it is set as "○".

Further, in parallel with the processing of STEP 70 and 80, the customer estimation device 1 assigns a score for each ID of the database holding the indexed item (STEP 75), and based on the score, sets the appearance rate (step S60). In consideration, "○" or "×" is given (STEP85). This result is used to verify the reliability.

After that, the customer estimation device 1 verifies the reliability of the determination (execution result) (STEP 90). Specifically, the reliability is verified by comparing the "answer" in the database holding the indexed items with the result of the "judgment" obtained in STEP85. This completes a series of processes for acquiring the target customer information.

The contents of the list, the score calculation method, and the like in the above-described embodiment are merely examples, and can be appropriately changed according to the intended use, purpose, and the like. Further, in FIG. 6, the reliability is verified in parallel with the actual execution, but the reliability may be acquired in the pre-process or the post-process of the actual execution.

Further, in the above-described embodiment, the company X that provides the service owns the source data S, and the client company Y provides the customer-owned items to the company X. However, the company Y may own the source data Sy and use it. Further, it is conceivable that Company Y can use the source data Sz owned by another company (for example, Company X).

The item correspondence unit 4 of the customer estimation device 1 has shown an example of adjusting so as to match the item of the higher-level concept, but it is also effective to adopt the one with the newer freshness of the item.

In the above embodiment, an example is shown in which a customer with a final score of "high" should be approached, but a customer with a "low" score also finds value in "low" and another approach. Can be the target of. Customers with a "low" score may also be excluded from the approach and used as a cut-off.

1 ... Customer estimation device, 2 ... Indexed item acquisition unit, 3 ... Customer owned item acquisition unit, 4 ... Item correspondence unit, 5 ... Classification item creation unit (machine learning model), 6 ... Score giving unit, 7 ... Appearance rate Estimating unit, 8 ... Customer judgment unit, 9 ... Reliability verification unit, 15 ... List, 16 ... Customer list, 17 ... Information list, 20 ... Customer DB, 30 ... Information DB.

Claims (6)

Indexed item acquisition means to acquire indexed items to be indexed,
Customer-owned item acquisition means for acquiring customer-owned items owned by the customer,
Obtained from the customer-owned item acquired by the customer-owned item acquisition means, a survey conducted on a specific or unspecified person, or history information of the person, and at least the attributes, values, tastes, and consumption of the person. Item correspondence means for associating the source data item of the source data in which either the action or the action pattern is reflected, and the item correspondence means.
A machine learning model in which the source data item corresponding to the customer-owned item is machine-learned as input data,
A classification item creation means for classifying the customer-owned items, which are elements necessary for estimating a target customer who will be in demand in the future, using the machine learning model according to the contents of the indexed items, and creating the classification items. When,
A scoring means for assigning a score to the classification item created by the classification item creating means based on a predetermined rule using the machine learning model, and
By referring to the information database to hold information related to the indexing item, the occurrence rate estimating means for estimating the incidence of the target customer for the indexed item,
The incidence the estimating the number of target customers from the estimating means the incidence estimated the total number of customers, higher Granted the score for each item of data groups having a reply to the indexed item A customer determination means for determining the target customer by selecting from
A customer estimation device characterized by comprising. The first aspect of claim 1, wherein the reliability verification means for verifying the reliability calculated from the degree of agreement obtained by comparing the answer with the result of the target customer determined by the customer determination means is provided. Customer estimator. When the number of customers having the same score exceeds the number of customers corresponding to the appearance rate due to the existence of a plurality of the customers having the same score on the determination line score determined from the appearance rate, the customer determination means arbitrarily extracts from the customers and the target. The customer estimation device according to claim 1 or 2, wherein the number of customers is adjusted. A customer estimation method that estimates a target customer who is interested in a product or service using a computer equipped with a machine learning model.
The indexing item acquisition step in which the data acquisition unit of the computer acquires the indexing item to be indexed,
The customer-owned item acquisition step in which the data acquisition unit acquires the customer-owned item owned by the customer,
The data processing unit of the computer is acquired from the customer-owned item acquired in the customer-owned item acquisition step, a survey conducted on a specific or unspecified person, or the history information of the person, and at least that person. Item correspondence steps that correspond to the source data items of the source data that reflect any of attributes, values, preferences, consumption behavior, and behavior patterns, and
A machine learning model creation step of creating a machine learning model by machine learning the source data item corresponding to the customer-owned item as input data.
The data processing unit classifies the customer-owned items, which are elements necessary for estimating the target customer who will be in demand in the future, using the machine learning model according to the contents of the indexed items, and classifies the classified items. The classification item creation step to be created and
A score-giving step in which the data processing unit assigns a score to the classification item created in the classification item creation step based on a predetermined rule using the machine learning model.
Wherein the data processing unit, by referring to the information database to hold information related to the indexing item, the appearance rate estimation step of estimating the incidence of the target customer for the indexed item,
Wherein the data processing unit, wherein the estimating the number of target customers, granted to each item of the data groups having a reply to the indexed item from the appearance ratio and the total number of customers estimated by the incidence estimation step A customer determination step of selecting from the highest score and determining the target customer, and
A customer estimation method characterized by comprising. The data processing unit includes a reliability verification step for verifying the reliability calculated from the degree of agreement obtained by comparing the answer with the result of the target customer determined in the customer determination step. The customer estimation method according to claim 4. In the customer determination step, when the number of customers having the same score exceeds the number corresponding to the appearance rate due to the existence of a plurality of the customers having the same score on the determination line score determined from the appearance rate, the target is arbitrarily extracted from the customers. The customer estimation method according to claim 4 or 5 , wherein the number of customers is adjusted.

Images