JP2014109898A - Analysis program, analysis method, and analysis device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To facilitate an analysis of the purchase situation by separating an article different in purchasing time as a different article even in the case that the same article is purchased.SOLUTION: A purchaser who purchases merchandise in the morning time zone, a purchaser who purchases merchandise in the time zone from the early-afternoon to the evening, and a purchaser who purchases merchandise in the time zone from the early-evening to the night are highly likely to have a different life style, and likely to have a different purchase behavior. An analysis device 100 clusters a plurality of purchasers with an element including a pair of at least identification information of merchandise and purchase time information as a clustering unit. The purchase time information, for example, means information showing a purchase time zone defined by a purchase time. Though a purchaser x and a purchaser y purchase the same merchandise, they are not classified into the same group since there is no commonness found between both purchasers when clustered as a pair.

Description

  The present invention relates to an analysis program, an analysis method, and an analysis apparatus.

  Conventionally, there is a clustering technique for classifying data into several groups. For example, a technique for investigating purchase behavior of a purchaser in purchasing at a store using clustering is known (see, for example, Patent Documents 1 to 3 below).

JP 2000-285175 A JP 09-305571 A JP 2001-134648 A

  However, conventionally, since there are various attributes even among purchasers of the same product, it is difficult to analyze the purchase status of the purchaser.

  In one aspect, an object of the present invention is to make it possible to analyze a purchase situation by separating products having different purchase times even as the same product as different products.

  According to one aspect of the present invention, referring to the storage content of a storage unit that stores product purchase history information including product identification information and purchase time information of the product for a plurality of products, at least product identification information; An analysis program, an analysis method, and an analysis apparatus that cluster an element including purchase time information as a clustering unit are proposed.

  According to one aspect of the present invention, it is possible to analyze a purchase situation by separating products with different purchase times even as the same product as different products.

FIG. 1 is an explanatory diagram illustrating an operation example of the analysis apparatus. FIG. 2 is an explanatory diagram illustrating an application example of the analysis apparatus. FIG. 3 is a block diagram of a hardware configuration example of the analysis apparatus according to the embodiment. FIG. 4 is a block diagram showing a functional configuration of the analyzer. FIG. 5 is an explanatory diagram showing an example of a purchase time division table. FIG. 6 is an explanatory diagram showing an example of purchase history information. FIG. 7 is an explanatory diagram illustrating an example of a combination of a product name and purchase time information. FIG. 8 is an explanatory diagram of the clustering result according to the first embodiment. FIG. 9 is a flowchart of a processing procedure example performed by the analyzer according to the first embodiment. FIG. 10 is an explanatory diagram illustrating an example of a combination of a product name and information indicating a specific date. FIG. 11 is an explanatory diagram of the clustering result according to the second embodiment. FIG. 12 is a flowchart illustrating an example of a processing procedure performed by the analysis apparatus according to the second embodiment. FIG. 13 is an explanatory diagram illustrating an example of a time interval division table. FIG. 14 is an explanatory diagram illustrating an example of a combination of information indicating a product name, a time zone, and a time interval. FIG. 15 is an explanatory diagram of the clustering result according to the third embodiment. FIG. 16 is a flowchart (part 1) illustrating an example of a processing procedure performed by the analysis apparatus according to the third embodiment. FIG. 17 is a flowchart (part 2) illustrating an example of a processing procedure performed by the analysis apparatus according to the third embodiment.

  Exemplary embodiments of an analysis program, an analysis method, and an analysis apparatus according to the present invention will be described below in detail with reference to the accompanying drawings. In the present embodiment, an example in which the analysis apparatus clusters a plurality of purchasers will be described.

  FIG. 1 is an explanatory diagram illustrating an operation example of the analysis apparatus. The analysis device 100 is a computer that analyzes purchase behavior of a purchaser in purchase at a store. As shown in the left side of FIG. 1, when clustering is performed only by the commonality of the products purchased by the purchaser, the purchaser x and the purchaser y purchase the same product, and therefore clustering into the same group. Is done.

  As described above, there are various attributes even among purchasers of the same product. For example, in a store like a department store or a supermarket, a purchaser who purchases a product in the morning, a purchaser who purchases a product from the early afternoon in the afternoon, and a purchase that purchases a product in the evening from the afternoon. There is a high possibility that the lifestyles will be different. Therefore, purchasing behavior may be different. For example, a purchaser who purchases a product during the morning hours is a person who does not work in the morning, but a purchaser who purchases a product from the evening in the afternoon to the evening time works until the evening. It is possible to guess that the person is.

  Therefore, in the present embodiment, the analysis apparatus 100 clusters a plurality of purchasers using a clustering unit as an element including at least product identification information and purchase time information. The purchase time information is information indicating a purchase time zone determined by the purchase time, for example. The purchase time zone is determined in advance by an analyst or the like based on a business form such as a store to be analyzed. The element indicates, for example, purchase contents including a combination of at least product identification information and purchase time information, and identification information indicating a purchaser. Thereby, according to the analysis apparatus 100, even if it is the same goods, it becomes possible to isolate | separate goods with different purchase times as different goods, and to analyze a purchase condition. In the example of FIG. 1, the purchaser x and the purchaser y purchase the same product, but when clustered according to the set, there is no commonality and therefore they are not classified into the same group. Thereby, even if it is purchasers of the same goods, if a purchase time slot | zone differs, it can classify as a purchaser which has another characteristic. Therefore, the purchaser can be distinguished according to the difference in lifestyle, and the purchase behavior of the purchaser can be easily analyzed.

  Further, the analyst may compare the number of persons belonging to the groups, or may perform detailed analysis for each group.

  FIG. 2 is an explanatory diagram illustrating an application example of the analysis apparatus. Purchase history information 600 indicating the purchase history of the purchaser can be obtained by sales using the Internet or a television or sales using an actual store. The analysis apparatus 100 clusters a plurality of purchasers based on the purchase history information 600. For example, the analysis apparatus 100 may output the clustering result to an apparatus that can be operated by the analyst. The analyst can analyze the purchase behavior tendency of each purchaser based on the clustering result.

(Hardware configuration example of analyzer 100)
FIG. 3 is a block diagram of a hardware configuration example of the analysis apparatus according to the embodiment. In FIG. 3, the analysis apparatus 100 includes a CPU (Central Processing Unit) 301, a ROM (Read-Only Memory) 302, a RAM (Random Access Memory) 303, a disk drive 304, a disk 305, and a network I / F. (InterFace) 306. Each unit is connected by a bus 300.

  Here, the CPU 301 controls the entire analysis apparatus 100. The ROM 302 stores a program such as a boot program. The RAM 303 is used as a work area for the CPU 301. The disk drive 304 controls reading / writing of data with respect to the disk 305 according to the control of the CPU 301. The disk 305 stores data written under the control of the disk drive 304.

  The network I / F 306 is connected to a network NET such as a LAN (Local Area Network), a WAN (Wide Area Network), or the Internet through a communication line, and is connected to another network 200 such as a device 200 that can be operated by an analyst via the network NET. Connected to the device. The network I / F 306 controls an internal interface with the network NET, and controls data input / output from an external device. For example, a modem or a LAN adapter may be employed as the network I / F 306.

  Further, although the analysis apparatus 100 gives an example in which the performance of a processor such as a server is high, the present invention is not limited thereto, and may be a PC (Personal Computer) or the like. Therefore, the analysis apparatus 100 may include an input device such as a keyboard and a mouse and an output device such as a display.

(Functional configuration example of the analyzer 100)
FIG. 4 is a block diagram showing a functional configuration of the analyzer. The analysis apparatus 100 includes a storage unit 401, a creation unit 402, and a clustering unit 403. The storage unit 401 is realized by a storage device such as the RAM 303 and the disk 305, for example. The processes of the creation unit 402 and the clustering unit 403 are coded in an analysis program stored in a storage device accessible by the CPU 301, for example. Then, the CPU 301 reads the analysis program from the storage device and executes the processing coded in the clustering program. Thereby, the process of each part is implement | achieved. Further, the processing results of the respective units are stored in a storage device such as the RAM 303 and the disk 305, for example.

  Here, in the present embodiment, the creation unit 402, the storage unit 401, and the clustering unit 403 will be described in detail using the first to third embodiments. In the first embodiment, a combination of product identification information and purchase time information is clustered as a clustering element. In the second embodiment, a combination of product identification information, purchase time information, and information indicating a purchase date is clustered as a clustering element. In the third embodiment, a combination of product identification information, purchase time information, and purchase time intervals by the same purchaser is clustered as a clustering element.

Example 1
In the first embodiment, a combination of product identification information and purchase time information is clustered as a clustering element. Thereby, even if it is the same product, it is possible to separate the products with different purchase times as different products and analyze the purchase situation. The purchase time zone is used as the purchase time information, and the purchase time zone classification is defined in the purchase time division table. FIG. 5 shows an example of a purchase time division table, and FIG. 6 shows a detailed example of the purchase history information 600 shown in FIG.

  FIG. 5 is an explanatory diagram showing an example of a purchase time division table. The purchase time division table 500 has information indicating a time zone, for example. According to the purchase time division table 500, each time from “10:00 to 20:00” is divided into three time zones. A time zone of “10: 0 to 12:59” represented by “AM”, a time zone of “13: 0 to 16:59” represented by “PM”, and “ZM” represented by “ZM” 17:00 to 20:00 ". As described above, the purchase time zone is determined in advance by an analyst or the like based on a business form such as a store to be analyzed. The purchase time division table 500 is realized by a storage device such as the RAM 303 and the disk 305, for example.

  FIG. 6 is an explanatory diagram showing an example of purchase history information. The purchase history information 600 includes a store name, a product name, a purchase date, a purchase time, a membership number, and other fields. A record is stored by setting information in each field. The purchase history information 600 is stored in the storage unit 401.

  Identification information indicating a store to be analyzed is registered in the store name field. In the product name field, the product name is registered as identification information of the purchased product. Here, the product identification information is used as the product name. However, the present invention is not limited to this, and may be, for example, a number assigned to the product. In the purchase date field, the date on which the product is purchased is registered. In the purchase time field, the time when the product is purchased is registered. In the member number field, the member number of the store to be analyzed is registered as purchaser identification information. Here, the purchaser's identification information is used as the membership number. However, the present invention is not limited to this. For example, the purchaser's name may be used. Other information other than the product name and membership number is registered in the other fields. For example, the other information includes the sex of the purchaser and the age of the purchaser, but is not particularly limited here.

  The creation unit 402 identifies the purchase time zone of the product based on the purchase time division table 500 from the purchase time of the product included in the purchase history information 600 for each of the purchaser identification information, and stores the product name and purchase in the storage unit 401. A combination with information indicating a time zone is stored. This set is the purchase content. For example, the identification information of the purchaser is stored in association with ““ product name ”+“ _ ”+“ information indicating the purchase time zone ””.

  FIG. 7 is an explanatory diagram illustrating an example of a combination of a product name and purchase time information. The stored content 700 includes, for example, ““ product name ”+“ _ ”+“ information indicating a purchase time zone ”” for each member number that is identification information of the purchaser. For example, since the purchase time of the product A is “10:30” for the member number “001”, the creation unit 402 specifies that the purchase was made during the “AM” time zone, and the member numbers “001” and “A_AM”. Is stored in the storage unit 401 in association with each other. Then, for example, for the member number “001”, the creation unit 402 specifies that the purchase time of the product B is “10:45”, so that it was purchased in the “AM” time zone, and the member number “001” is obtained. “B_AM” is associated and stored in the storage unit 401. The creation unit 402 also specifies the purchase time zone from the purchase time of the purchased products for other member numbers in the same manner, and stores the purchase time zone in the storage unit 401.

  Then, the clustering unit 403 refers to the storage contents of the storage unit 401 and clusters the elements including at least the product identification information and the purchase time information as a clustering unit. Specifically, the clustering unit 403 clusters a plurality of purchasers using a combination of a product name stored in association with each membership number in the storage content 700 of the storage unit 401 and information indicating a purchase time zone as a clustering unit. . A detailed method of clustering is not particularly limited.

  FIG. 8 is an explanatory diagram of the clustering result according to the first embodiment. FIG. 8 shows an example in which each member number is clustered into one of group G1 to group Gn (n ≧ 1) and the number of purchasers in each group is graphed.

  The group G1 includes a membership number “001” and a membership number “002” having purchase contents “A_AM” and “B_AM”. The group G2 includes a membership number “003”, a membership number “004”, a membership number “007”, and a membership number “008” with purchase contents “A_PM” and “B_PM”. A membership number “005” and a membership number “006” having purchase contents “C_ZM” and “D_ZM” belong to the group G3.

  When clustering is performed using only the product identification information as an element, the group G1 and the group G2 are the same group. However, according to the first embodiment, the identification information of purchasers having different purchase time zones is set to different groups G1. And group G2. Therefore, it is possible to analyze a purchase situation by separating products having different purchase times even as the same product as different products.

  For example, an analyst may analyze purchase behavior in detail about each of groups G1-Gn. Specifically, for example, the analyst analyzes the purchasers who purchase the products A and B in the morning by analyzing the sex and occupation of the purchaser whose membership number is included in the group G1. May be. Further, for example, the analyst may analyze the purchasing behavior of the entire purchaser by comparing the groups G1 to Gn.

(Example of processing procedure performed by the analyzer 100 according to the first embodiment)
FIG. 9 is a flowchart of a processing procedure example performed by the analyzer according to the first embodiment. First, the analysis apparatus 100 sets a purchase time section (step S901), acquires purchase history information 600 (step S902), and determines whether there is an unselected member number (step S903). When there is an unselected member number (step S903: Yes), the analysis apparatus 100 selects one member number from the unselected member numbers (step S904).

  The analysis apparatus 100 acquires a record including the selected membership number from the purchase history information 600 (step S905), and determines whether there is an unselected record among the acquired records (step S906). When there is no unselected record among the acquired records (step S906: No), the process returns to step S903. When there is an unselected record among the acquired records (step S906: Yes), the analysis apparatus 100 selects one record from the unselected records (step S907), and from the acquired record “product name”, “Purchase time” is acquired (step S908). Then, the analysis apparatus 100 determines which of the purchase time segments the purchase time zone is (step S909).

  When the time zone of purchase time is “AM” (step S909: AM), the analysis apparatus 100 sets purchase content = “product name” + “_” + “AM” (step S910), and proceeds to step S913. . When the time zone of the purchase time is “PM” (step S909: PM), the analysis apparatus 100 sets purchase content = “product name” + “_” + “PM” (step S911), and proceeds to step S913. .

  When the time zone of the purchase time is “ZM” (step S909: ZM), the analysis apparatus 100 sets purchase content = “product name” + “_” + “ZM” (step S912), and proceeds to step S913. . After step S910, step S911, or step S912, the analysis apparatus 100 stores the selected member number and purchase content in association with each other in the storage unit 401 (step S913), and returns to step S906.

  When there is no unselected membership number (step S903: No), the analysis apparatus 100 refers to the storage content 700 of the storage unit 401 and clusters the elements including the purchase content as a clustering unit (step S914), and the clustering result Is output (step S915).

(Example 2)
In the second embodiment, a plurality of purchasers are clustered using a combination of product identification information, product purchase time information, and information based on the purchase date of the product as a clustering element. As a result, even if the purchased product and the purchase time zone are the same, it is possible to analyze the purchase situation by separating the product which is different in the day of the week, weekday or holiday as a different product. The purchase history information 600 and the purchase time division table 500 will be described using the same example as in the first embodiment.

  The creation unit 402 determines whether the purchase date of the product included in the purchase history information 600 is a specific date for each piece of purchaser identification information. The specific day may be, for example, a holiday or a weekday, a specific day of the week, or a specific date. Here, a specific day is described as a holiday. When the purchase date of the product included in the purchase history information 600 is a holiday, the creation unit 402 stores the product name and information indicating the holiday in the storage unit 401. For example, the member number of the purchaser is associated with ““ product name ”+“ _ ”+“ holiday information ”” and stored in the storage unit 401.

  When the purchase date of the product included in the purchase history information 600 is a weekday, the creation unit 402 selects the product based on the purchase time division table 500 from the purchase time of the product included in the purchase history information 600 as in the first embodiment. The purchase time zone is specified and stored in the storage unit 401. For example, the purchaser's membership number is associated with ““ product name ”+“ _ ”+“ information indicating a time zone ”” and stored in the storage unit 401.

  FIG. 10 is an explanatory diagram illustrating an example of a combination of a product name and information indicating a specific date. The stored content 1000 includes, for example, ““ product name ”+“ _ ”+“ information indicating a time zone ”” or ““ product name ”+“ _ ”+“ holiday ”for each member number that is identification information of the purchaser. “” Indicating “”. For example, for the member number “003”, the creation unit 402 specifies that the purchase date of the product A is a holiday, and stores the member number “003” and “A_HOL” in the storage unit 401 in association with each other. For example, for the member number “003”, the creation unit 402 specifies that the purchase date of the product B is a holiday, and stores the member number “003” and “B_HOL” in the storage unit 401 in association with each other. Therefore, “003 A_HOL B_HOL” is stored in the storage unit 401.

  Also for the member number “004”, the creation unit 402 specifies that the purchase date of the products A and B is a holiday, associates the member number “004” with the purchase contents “A_HOL” and “B_HOL”, and stores them. 401 is stored. Therefore, “004 A_HOL B_HOL” is stored in the storage unit 401. Member numbers “001”, “002”, “005”, “006”, “007”, and “008” are weekdays and are the same as in the first embodiment.

  The clustering unit 403 refers to the storage contents of the storage unit 401. When the purchase date is a specific date, the clustering unit 403 includes an element including a combination of product identification information and information indicating the specific date in a clustering unit. Cluster as Further, when the purchase date is not a specific date, the clustering unit 403 performs clustering using elements including the product identification information and the purchase time information as a clustering unit, as in the first embodiment. The clustering unit 403 uses a combination of a product name stored in association with each membership number of the storage content 1000 in the storage unit 401 and information indicating a purchase time zone or information indicating a purchase date as a clustering unit. Cluster membership numbers into multiple groups.

  FIG. 11 is an explanatory diagram of the clustering result according to the second embodiment. FIG. 11 shows an example in which each membership number is clustered into one of group G1 to group Gm (m ≧ 1), and the number of purchasers is graphed for each group.

  The group G1 includes a membership number “001” and a membership number “002” having purchase contents “A_AM” and “B_AM”. Member numbers “003” and “004” having purchase contents “A_HOL” and “B_HOL” belong to the group G2. Member numbers “005” and “006” having purchase contents “C_ZM” and “D_ZM” belong to the group G3. Member numbers “007” and “008” having “A_PM” and “B_PM” as elements belong to the group G4.

  According to the second embodiment, even when the purchased product and the purchase time zone are the same, it is possible to analyze a purchase situation by separating a product with different day of the week, weekday, or holiday as a different product. . In the first embodiment, the member numbers “003”, “004”, “007”, and “008” are classified into the same group. In the second embodiment, the member numbers “003”, “004”, and the member number “ “007” and “008” are classified into different groups. For example, there is a high possibility that lifestyles are different between those who shop on weekdays and those who shop on holidays. Therefore, even for purchasers of the same product, the purchase status is analyzed as a purchaser with different characteristics depending on whether the purchase date is a holiday or not, and the lifestyle differences of each purchaser are taken into account. Enables analysis of purchasing status.

  For example, an analyst may analyze purchase behavior in detail about each of groups G1-Gm. Specifically, for example, the analyst analyzes the purchasers who purchase the products A and B in the morning by analyzing the sex and occupation of the purchaser whose membership number is included in the group G1. May be. Further, for example, the analyst may analyze the purchasing behavior of the entire purchaser by comparing the groups G1 to Gm.

(Example of processing procedure performed by the analyzer 100 according to the second embodiment)
FIG. 12 is a flowchart illustrating an example of a processing procedure performed by the analysis apparatus according to the second embodiment. First, the analysis apparatus 100 sets a purchase time segment (step S1201). Next, the analysis apparatus 100 acquires the purchase history information 600 (step S1202), and determines whether there is an unselected member number (step S1203). When there is an unselected member number (step S1203: Yes), the analysis apparatus 100 selects one member number from the unselected member numbers (step S1204).

  The analysis apparatus 100 acquires a record including the selected membership number from the purchase history information 600 (step S1205), and determines whether there is an unselected record among the acquired records (step S1206). When there is no unselected record among the acquired records (step S1206: No), the process returns to step S1203. When there is an unselected record among the acquired records (step S1206: Yes), the analysis apparatus 100 selects one record from the unselected records (step S1207), and from the acquired record, “product name”, “Purchase date” and “purchase time” are acquired (step S1208).

  The analysis apparatus 100 determines whether the “purchase date” is a holiday or a weekday (step S1209). When it is a weekday (step S1209: weekday), the analysis apparatus 100 determines whether the time zone of the purchase time is the purchase time segment (step S1210). When the time zone of purchase time is “AM” (step S1210: AM), the analysis apparatus 100 sets purchase content = “product name” + “_” + “AM” (step S1211), and proceeds to step S1215. . When the time zone of the purchase time is “PM” (step S1210: PM), the analysis apparatus 100 sets purchase content = “product name” + “_” + “PM” (step S1212), and proceeds to step S1215. .

  When the time zone of the purchase time is “ZM” (step S1210: ZM), the analysis apparatus 100 sets purchase content = “product name” + “_” + “ZM” (step S1213), and proceeds to step S1215. . If it is determined in step S1209 that it is a holiday (step S1209: holiday), the analysis apparatus 100 sets purchase content = “product name” + “_” + “HOL” (step S1214), and proceeds to step S1215. .

  After step S1211, step S1212, step S1213, or step S1214, the analysis apparatus 100 stores the selected member number and the purchase content in association with each other in the storage unit 401 (step S1215), and proceeds to step S1206. .

  In step S1203, when there is no unselected membership number (step S1203: No), the analysis apparatus 100 refers to the storage content 1000 of the storage unit 401 and clusters the elements including the purchase content as a clustering unit (step S1216). ), And outputs the clustering result (step S1217).

(Example 3)
In the third embodiment, clustering is performed using elements including a combination of product identification information, product purchase time information, and product purchase time interval as a clustering unit. Thereby, even if the purchased product and the purchase time zone are the same, it is possible to separate the products having different purchase time intervals as different products and analyze the purchase situation. The purchase history information 600 and the purchase time division table 500 will be described using the same example as in the first embodiment. In the present embodiment, the information indicating the purchase time interval is determined by the time interval division based on the time interval division table. The time interval division table is shown in FIG.

  FIG. 13 is an explanatory diagram illustrating an example of a time interval division table. According to the time interval division table 1300, each time interval is classified into “0.5 hours or less”, “longer than 0.5 hours, less than 3 hours”, and “3 hours or more”. When the time interval is “0.5 hours or less”, it is represented by “Short”, and when the time interval is “greater than 0.5 hour and less than 3 hours”, it is represented by “Middle” and the time interval is “3”. “More than time” is represented by “Long”. The time interval division is not limited to the example of FIG. 13 and can be variously changed. The time interval division table 1300 is realized by a storage device such as the RAM 303 and the disk 305, for example.

  The creation unit 402 identifies the time interval between the purchase dates and times of the two purchased products included in the purchase history information 600 for each of the purchaser identification information. The creation unit 402 identifies a time interval segment based on the time interval segment table 1300 from the identified time interval. Then, the creation unit 402 specifies the purchase time zone of the product based on the purchase time division table 500 from the purchase time of the product included in the purchase history information 600 for each of the two products, as in the first embodiment. The creation unit 402 stores the identification information of the purchaser, the identification information of the two products, the purchase time zone of the two products, and the identified time interval division in the storage unit 401 in association with each other.

  For example, for each member number of the purchaser, for example, ““ first product name ”+“ _ ”+“ second product name ”+“ _ ”+“ information indicating the first time zone ”+“ The member number of the purchaser is associated with “_” + “information indicating the second time zone” + “_” + “time interval” and stored in the storage unit 401.

  FIG. 14 is an explanatory diagram illustrating an example of a combination of information indicating a product name, a time zone, and a time interval. The stored content 1400 includes, for example, ““ first product name ”+“ _ ”+“ second product name ”+“ _ ”+“ information indicating the first time zone ”+“ _ ”for each member number. “+“ Information indicating the second time zone ”+“ _ ”+“ time interval ”” is included.

  For example, since the purchase time of the product A is “10:30” for the member number “001”, the creation unit 402 specifies that the purchase was made during the “AM” time zone, and the purchase time of the product B is “10”. : 45 ”, it is specified that the purchase was made in the“ AM ”time zone. For example, the creation unit 402 specifies the time interval between the purchase date and time of the product A and the purchase date and time of the product B for the member number “001”. This time interval is 8 days and 15 minutes. The creation unit 402 identifies a time interval segment based on the time interval segment table 1300 from the identified time interval. The time interval division here is “Long” because it is “3 hours or more”. Then, the creation unit 402 includes a membership number “001”, a first product name “A”, a second product name “B”, a first time zone “AM”, and a second time zone “AM”. The time interval section “Long” is associated with each other and stored in the purchase history information 1400. Therefore, “001 A_B_AM_AM_Long” is stored in the storage unit 401. The creation unit 402 performs the same process for other member numbers.

  As for the time interval between purchases by the same consumer, for example, all the intervals between the closest dates may be included as clustering elements, or only representative values such as average values are included as clustering elements. There is no particular limitation.

  Next, the clustering unit 403 refers to the storage content 1400 of the storage unit 401 and combines product identification information, product purchase time information, and information based on product purchase time intervals by the same purchaser. The elements included in are clustered as a clustering unit.

  FIG. 15 is an explanatory diagram of the clustering result according to the third embodiment. FIG. 15 shows an example in which each member number is clustered into one of group G1 to group Gl (l ≧ 1) and the number of purchasers is graphed for each group.

  Member numbers “001” and “002” having “A_B_AM_AM_Long” as purchase contents belong to the group G1. Member numbers “003” and “004” having purchase contents “A_B_PM_PM_Short” belong to the group G2. Member numbers “005” and “006” having “C_D_ZM_ZM_Long” as purchase contents belong to the group G3. A membership number “007” and a membership number “008” whose purchase contents are “A_B_PM_PM_Middle” belong to the group G4.

  According to the third embodiment, even when the purchased product and the purchase time zone are the same, it is possible to analyze the purchase situation by separating the products having different purchase time intervals as different products. In the first embodiment, the member numbers “003”, “004”, “007”, “008” are classified into the same group, but the member numbers “003”, “004”, the member numbers “007”, “ “008” are classified into different groups.

  Further, for example, the analyst may analyze purchase behavior in more detail for each of the groups G1 to Gl. Specifically, for example, the analyst analyzes the purchasers who purchase the products A and B in the morning by analyzing the sex and occupation of the purchaser whose membership number is included in the group G1. May be. Further, for example, the analyst may analyze the purchasing behavior of the entire purchaser by comparing the groups G1 to G1.

  Moreover, about Example 2 and Example 3, various changes, such as combining, are possible.

(Example of processing procedure performed by the analyzer 100 according to the third embodiment)
16 and 17 are flowcharts illustrating an example of a processing procedure performed by the analyzer according to the third embodiment. First, the analysis apparatus 100 sets a purchase time segment and a time interval segment (step S1601), acquires purchase history information 600 (step S1602), and determines whether there is an unselected member number (step S1603). ). When there is an unselected member number (step S1603: Yes), the analysis apparatus 100 selects one member number from the unselected member numbers (step S1604).

  The analysis apparatus 100 acquires a record including the selected membership number from the purchase history information 600 (step S1605), and determines whether there is an unselected record among the acquired records (step S1606). If there is no unselected record among the acquired records (step S1606: No), the process returns to step S1603. When there is an unselected record among the acquired records (step S1606: Yes), the analysis apparatus 100 selects a record with the earliest purchase date and purchase time from the unselected records (step S1607), and the acquired record "Product name", "Purchase time", and "Purchase date" are acquired from (S1608). The analysis apparatus 100 selects the second record whose purchase date and time is closest to the purchase date and time of the first record from the unselected records (step S1609).

  The analysis apparatus 100 acquires “product name”, “purchase time”, and “purchase date” from the selected second record (step S1610). The analysis apparatus 100 sets purchase content = “product name” of the first record + “_” + “product name” of the second record (step S1611).

  Next, the analysis apparatus 100 determines which of the purchase time sections the time zone of the purchase time of the first record (step S1701). When the time zone of the purchase time is “AM” (step S1701: AM), the analysis apparatus 100 sets purchase content = purchase content + “_” + “AM” (step S1702), and proceeds to step S1705.

  When the time zone of the purchase time is “PM” (step S1701: PM), the analysis apparatus 100 sets purchase content = purchase content + “_” + “PM” (step S1703), and proceeds to step S1705. When the time zone of the purchase time is “ZM” (step S1701: ZM), the analyzer 100 sets purchase content = purchase content + “_” + “ZM” (step S1704), and proceeds to step S1705.

  The analysis apparatus 100 determines which time zone of the purchase time of the second record is the purchase time segment (step S1705). When the time zone of the purchase time is “AM” (step S1705: AM), the analysis apparatus 100 sets purchase content = purchase content + “_” + “AM” (step S1706), and proceeds to step S1709.

  When the time zone of the purchase time is “PM” (step S1705: PM), the analysis apparatus 100 sets purchase content = purchase content + “_” + “PM” (step S1707), and proceeds to step S1709. When the time zone of the purchase time is “ZM” (step S1705: ZM), the analyzer 100 sets purchase content = purchase content + “_” + “ZM” (step S1708), and proceeds to step S1709.

  Next, the analysis apparatus 100 identifies the time interval between the purchase dates of the first and second records (step S1709), and determines how much the difference is (step S1710). In the case of 0.5 hours or less (step S1710: 0.5 or less), the analysis apparatus 100 sets purchase content = purchase content + “_” + “short” (step S1711), and proceeds to step S1714.

  When longer than 0.5 hours and less than 3 hours (step S1710: longer than 0.5 and less than 3), the analysis apparatus 100 sets purchase content = purchase content + “_” + “Middle” (step S1712), and step S1714. Migrate to In the case of 3 hours or more (step S1710: 3 or more), the analysis apparatus 100 sets purchase content = purchase content + “_” + “Long” (step S1713), and proceeds to step S1714.

  Next, the analysis apparatus 100 stores the selected member number and purchase contents in association with each other in the storage unit 401 (step S1714), and returns to step S1606.

  In step S1603, if there is no unselected membership number (step S1603: No), the analysis apparatus 100 refers to the storage content 1400 of the storage unit 401 and clusters the elements including the purchase content as a clustering unit (step S1612). ), And outputs the clustering result (step S1613).

  As described above, the analysis apparatus clusters the combination of the product identification information and the purchase time information as a clustering element. Further, the analysis apparatus clusters a plurality of purchasers. Thereby, even if it is purchasers of the same goods, if a purchase time slot | zone differs, it can classify as a purchaser which has another characteristic. Therefore, the purchaser can be distinguished according to the difference in lifestyle, and the purchase behavior of the purchaser can be easily analyzed.

  Further, the analysis device clusters the combination of the product identification information and the specific date as an element of clustering when the purchase date of the product is a specific date. Thereby, even if it is a purchaser of the same goods, if it is a weekday or a holiday, for example, it can classify | categorize as a purchaser which has a different characteristic. For example, a purchaser who purchases a product on weekdays and a purchaser who purchases a product on a holiday may have different lifestyles such as different occupations. Therefore, the purchaser can be distinguished according to the difference in lifestyle, and the purchase behavior of the purchaser can be easily analyzed.

  In addition, the analysis apparatus performs clustering using, as a clustering unit, elements including a combination of product identification information, product purchase time information, and purchase time intervals by the same purchaser. Thereby, even a purchaser who purchased the same product in the same time zone can be classified as a purchaser having different characteristics if the purchase time interval is different. For example, a purchaser with a short time interval can perform analysis such as purchasing behavior that purchases a product many times during one day.

  The analysis method described in the present embodiment can be realized by executing a prepared analysis program on a computer such as a PC or a workstation. The analysis program is recorded on a computer-readable recording medium such as a flash memory or a disk, and is executed by being read from the recording medium by the computer. The analysis program may be distributed via a network NET such as the Internet.

DESCRIPTION OF SYMBOLS 100 Analysis apparatus 401 Memory | storage part 402 Creation part 403 Clustering part 600 Purchase history information G1-Gn, Gm, Gl group

Claims (6)

A purchasing situation analysis program,
An element including at least product identification information and purchase time information in combination with reference to the storage content of a storage unit that stores product purchase history information including product identification information and purchase time information of the product for a plurality of products. Cluster as a clustering unit,
An analysis program for causing a computer to execute processing. The purchase history information includes, for each of a plurality of purchasers, identification information of the product purchased by the purchaser and purchase time information of the product,
The analysis program according to claim 1, wherein in the clustering process, the plurality of purchasers are clustered. The purchase history information includes purchase date information of the product,
In the clustering process, referring to the storage content of the storage unit, when the purchase date information of the product indicates a specific date, an element including at least the identification information of the product and the specific date in a set is included. The analysis program according to claim 1, wherein clustering is performed as a clustering unit.   In the process of clustering, referring to the storage content of the storage unit, at least an element including the identification information of the product, the purchase time information of the product, and the time interval of purchase by the same purchaser in a set The analysis program according to claim 2, wherein the plurality of purchasers are clustered as a clustering unit. A method for analyzing purchasing status,
An element including at least product identification information and purchase time information in combination with reference to the storage content of a storage unit that stores product purchase history information including product identification information and purchase time information of the product for a plurality of products. Cluster as a clustering unit,
An analysis method characterized in that the processing is executed by a computer. A purchasing status analyzer,
A storage unit for storing product purchase history information including product identification information and product purchase time information for a plurality of products;
A clustering unit that refers to the storage content of the storage unit and clusters the elements including at least product identification information and purchase time information as a clustering unit;
An analysis apparatus comprising:

Images