JP2012089014A - Purchase action analysis device, purchase action analysis method, and purchase action analysis program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To appropriately perform marketing analysis.SOLUTION: A purchase action analysis device obtains Web browsing data that includes Web pages browsed by a user at a user terminal. Then, the purchase action analysis device extracts page information that is information included in a Web page associated with a commodity out of Web pages browsed by the user by performing extraction processing on the obtained Web browsing data according to a predetermined extraction condition. After that, the purchase action analysis device extracts commodity information that is information associated with a commodity included in the page information, from the extracted page information. The purchase action analysis device generates summary information that is a summary of information included in the page information, from the extracted page information. Then, the purchase action analysis device performs predetermined statistical analysis processing on the commodity information and the summary information.

Description

  The present invention relates to a purchasing behavior analysis apparatus, a purchasing behavior analysis method, and a purchasing behavior analysis program.

  There is a technique for collecting data on consumer purchasing behavior and conducting marketing analysis based on the collected data. For example, there is an RFID technique in which a RFID (Radio Frequency IDentification) tag is attached to a product, and when a mobile terminal operated by a consumer contacts an article, the tag identifier is acquired and transmitted to a database. In the RFID technique, thereafter, the analysis device performs marketing analysis based on information stored in a database.

  In addition, for example, there is a method of performing marketing analysis using POS (Point Of Sale) system data, and there is a method of collecting data about users using a questionnaire and performing a marketing method. In addition, there is a method to analyze consumer leads and line of sight from surveillance cameras in the store, and to conduct marketing analysis based on the analysis results, and to analyze web logs obtained at specific EC (e-commerce) sites There is a technique for marketing analysis.

JP 2003-122896 A

  However, the above-described method has a problem that marketing analysis cannot be performed properly. For example, the data used in the method using POS does not include the behavior data before and after the purchase of the consumer, and the method using the questionnaire or the method using the monitoring camera collects the behavior data before and after the purchase of the consumer. Cost. In addition, for example, in the method using a questionnaire, there is a time lag between the time when a panel user takes a purchase action and the time when data is collected. Occurs. In addition, for example, in the method using a surveillance camera, only consumer behavior data in a specific store can be obtained, and behaviors comparing products and prices by consumers performed across other stores, etc. No behavior data can be obtained for.

  The disclosed technology has been made in view of the above, and an object thereof is to provide a purchasing behavior analysis apparatus, a purchasing behavior analysis method, and a purchasing behavior analysis program capable of appropriately executing marketing analysis.

  In one aspect, the disclosed purchasing behavior analysis apparatus includes an acquisition unit that acquires Web browsing data including a Web page viewed by a user on a user terminal. Further, the purchase behavior analysis device executes an extraction process on the Web browsing data acquired by the acquisition unit according to a predetermined extraction condition, so that the purchase behavior analyzing apparatus is included in the Web page related to the product among the Web pages browsed by the user. A page information extraction unit that extracts page information that is information is provided. In addition, the purchase behavior analysis apparatus includes a product information extraction unit that extracts product information that is information about the product included in the page information from the page information extracted by the page information extraction unit. In addition, the purchase behavior analysis apparatus includes a summary generation unit that generates summary information that is a summary of information included in the page information from the page information extracted by the page information extraction unit. In addition, the purchase behavior analysis device includes a statistical analysis unit that performs a predetermined statistical analysis process on the product information extracted by the product information extraction unit and the summary information generated by the summary generation unit.

  According to one aspect of the purchasing behavior analysis apparatus to be disclosed, there is an effect that marketing analysis can be appropriately executed.

FIG. 1 is a block diagram illustrating an example of the configuration of the purchase behavior analysis apparatus according to the first embodiment. FIG. 2 is a flowchart illustrating an example of a process flow performed by the purchase behavior analysis apparatus according to the first embodiment. FIG. 3 is a block diagram illustrating an example of the configuration of the purchase behavior analysis system according to the second embodiment. FIG. 4 is a diagram illustrating an example of information stored in the Web browsing data DB according to the second embodiment. FIG. 5 is a diagram illustrating an overview of processing by the purchase behavior analysis server in the second embodiment. FIG. 6 is a diagram illustrating an example of information stored in the text extraction rule DB according to the second embodiment. FIG. 7 is a diagram illustrating an example of information stored in the browse text data DB according to the second embodiment. FIG. 8 is a diagram illustrating an example of information stored in the browsing product data DB according to the second embodiment. FIG. 9 is a diagram illustrating an example of information stored in the purchased product data DB according to the second embodiment. FIG. 10 is a diagram illustrating an example of information stored in the behavior summary data DB according to the second embodiment. FIG. 11 is a diagram illustrating an example of marketing data stored in the marketing data DB according to the second embodiment. FIG. 12 is a diagram illustrating an example of marketing data stored in the marketing data DB according to the second embodiment. FIG. 13 is a flowchart illustrating an example of a process flow of the panel user terminal according to the second embodiment. FIG. 14 is a flowchart illustrating an example of a processing flow by the Web browsing data receiving unit according to the second embodiment. FIG. 15 is a flowchart illustrating an example of a registration process performed by the text extracting unit according to the second embodiment. FIG. 16 is a flowchart illustrating an example of the flow of extraction processing by the text extraction unit in the second embodiment. FIG. 17 is a flowchart illustrating an example of a process flow by the product information extraction unit according to the second embodiment. FIG. 18 is a flowchart illustrating an example of a processing flow by the summary generation unit according to the second embodiment. FIG. 19 is a flowchart illustrating an example of a process flow by the statistical analysis unit according to the second embodiment. FIG. 20 is a diagram for illustrating the effect of the second embodiment. FIG. 21 is a block diagram illustrating an example of the configuration of the purchasing behavior analysis system according to the third embodiment. FIG. 22 is a diagram illustrating an example of information stored in the personal information data DB according to the third embodiment. FIG. 23 is a flowchart illustrating processing by the panel user terminal according to the third embodiment. FIG. 24 is a diagram illustrating an example of a variation of the configuration of the purchase behavior analysis system. FIG. 25 is a diagram illustrating that information processing by the purchase behavior analysis program is specifically realized using a computer.

  Hereinafter, embodiments of a purchasing behavior analysis apparatus, a purchasing behavior analysis method, and a purchasing behavior analysis program to be disclosed will be described in detail based on the drawings. Note that the invention disclosed by this embodiment is not limited. Each embodiment can be appropriately combined within a range in which processing contents do not contradict each other.

[Configuration of Purchasing Behavior Analysis Device According to Embodiment 1]
An example of the configuration of the purchasing behavior analysis apparatus 100 according to the first embodiment will be described with reference to FIG. FIG. 1 is a block diagram illustrating an example of the configuration of the purchase behavior analysis apparatus according to the first embodiment. In the example illustrated in FIG. 1, the purchase behavior analysis apparatus 100 includes an acquisition unit 101, a page information extraction unit 102, a product information extraction unit 103, a summary generation unit 104, and a statistical analysis unit 105.

  The acquisition unit 101 acquires Web browsing data that is information about a Web page viewed by a panel user on the panel user terminal. Here, the web browsing data includes the contents of the web page browsed on the panel user terminal, for example, HTML (Hyper Text Markup Language) data of the web page.

  Here, the panel user terminal is also briefly described. The panel user terminal is a terminal used for a panel user to browse a web page. Examples of the panel user terminal include a known personal computer, a television set having a function of accessing the Internet, a mobile phone, a smartphone, a PHS (Personal Handy-phone System), a PDA (Personal Digital Assistant), and the like.

  Returning to the description of the acquisition unit 101. For example, the acquisition unit 101 acquires Web browsing data by inputting Web browsing data collected from a panel user terminal by an administrator of the purchase behavior analysis apparatus 100. For example, when the purchase behavior analysis apparatus 100 and the panel user terminal are connected, the acquisition unit 101 collects Web browsing data from the panel user terminal. The purchase behavior analysis apparatus 100 and the panel user terminal are connected, for example, via the Internet or connected using a dedicated line.

  The page information extraction unit 102 is information included in a Web page related to a product among Web pages browsed by the user by executing an extraction process on the Web browsing data acquired by the acquisition unit 101 according to a predetermined extraction condition. Extract some page information. For example, the page information extraction unit 102 executes the extraction process using a condition for specifying a website including information related to a product and a condition for specifying information extracted from the website as predetermined extraction conditions. Conditions for specifying a website that includes information about products are, for example, URLs (Uniform Resource Locators) of websites that contain information about products, phrases that are expected to be included in websites that contain information about products, and information about products For example, the title of a web page that is expected to be assigned in the case of a website including The condition for specifying the information extracted from the website corresponds to, for example, information for specifying the position in the website. However, the present invention is not limited to this, and any extraction condition may be used.

  For example, the page information extraction unit 102 extracts information that matches a condition for specifying information to be extracted from the Web site from Web browsing data for a Web page that matches a condition for specifying a Web site that includes information related to products. To explain with a more detailed example, the page information extraction unit 102 calculates the website from the web browsing data about the web page having the URL that matches the URL specified as the condition for identifying the website including the information about the product. By extracting the information at the position specified by the conditions for specifying the information to be extracted from, the product name, the website name, the text included in the website, the price notation, etc. are extracted.

  Note that the page information extracted by the page information extraction unit 102 includes not only product information that is information about products included in the page information, but also other information included in a website that includes information about products. . The information regarding the product is information indicating, for example, what product information the panel user has browsed from. For example, “Web site name”, “product name”, “price notation”, and the like correspond to product information, and “text” included in the Web site corresponds to other information.

  The product information extraction unit 103 extracts product information from the page information extracted by the page information extraction unit 102. That is, the page information extracted by the page information extraction unit 102 includes not only information related to products but also other information included in a website including information related to products. For example, the product information extraction unit 103 extracts “Web site name” as information indicating where the panel user has browsed, and “product” as information indicating what product information has been browsed. "Name" and "Price notation" are extracted.

  The summary generation unit 104 generates summary information, which is a summary of information included in the page information, from the page information extracted by the page information extraction unit 102. For example, the summary generation unit 104 generates summary information after using the product information and other information together. For the summary information generation processing by the summary generation unit 104, any method may be used. The summary information is also described as “behavior summary”.

  The statistical analysis unit 105 performs a predetermined statistical analysis process on the product information extracted by the product information extraction unit 103 and the summary information generated by the summary generation unit 104. That is, the statistical analysis unit 105 does not perform statistical analysis processing on the Web browsing data, but combines the data that is the processing result by the product information extraction unit 103 and the data that is the processing result by the summary generation unit 104. Statistical analysis processing is executed on the collected data.

[Processing of Purchasing Behavior Analysis Device According to Embodiment 1]
An example of the flow of processing by the purchasing behavior analysis apparatus 100 according to the first embodiment will be described with reference to FIG. FIG. 2 is a flowchart illustrating an example of a process flow performed by the purchase behavior analysis apparatus according to the first embodiment.

  As shown in FIG. 2, when the acquisition unit 101 acquires Web browsing data (Yes at Step S101), the page information extraction unit 102 extracts page information from Web browsing data (Step S102). That is, the page information extraction unit 102 performs extraction processing on Web browsing data according to a predetermined extraction condition, thereby obtaining page information that is information included in a Web page related to a product among Web pages browsed by a panel user. Extract. For example, the page information extraction unit 102 extracts a product name, a website name, a text included in the website, a price notation, and the like.

  Then, the product information extraction unit 103 extracts product information (step S103). For example, the product information extraction unit 103 extracts “Website name” from the page information as information indicating where the panel user has browsed, and “product name” as information indicating what product information has been browsed. Or "price notation".

  Then, the summary generation unit 104 generates summary information (step S104). For example, the summary generation unit 104 generates summary information after using the product information included in the page information and other information together. Then, the statistical analysis unit 105 executes product information, summary information, and predetermined statistical analysis processing (step S105).

  In addition, said process procedure is not limited to said order, You may change suitably in the range which does not contradict a process content. For example, the above step S103 may be executed after S104.

[Effect of Example 1]
As described above, according to the purchasing behavior analysis apparatus 100 according to the first embodiment, the web browsing data including the web page browsed by the panel user is acquired at the panel user terminal, and the web browsing data is obtained according to the predetermined extraction condition. By executing the extraction process, page information that is information included in the Web page related to the product is extracted from the Web pages browsed by the user. Then, the purchasing behavior analysis apparatus 100 extracts product information that is information about the product included in the page information from the page information, generates summary information that is a summary of the information included in the page information from the page information, and stores the product information A predetermined statistical analysis process is performed on the summary information. As a result, according to the purchasing behavior analysis apparatus 100 according to the first embodiment, marketing analysis can be appropriately executed.

  That is, according to the purchasing behavior analysis apparatus 100 according to the first embodiment, by using Web browsing data, it is possible to collect and use data on a large scale at a low cost. Moreover, according to the purchasing behavior analysis apparatus 100, by performing a statistical analysis process based on the web browsing data collected by the panel user terminal, the time between the time when the purchasing behavior occurs and the time when the data is collected Answer mistakes and omissions due to time lag can be prevented. Further, by executing statistical analysis processing based on Web browsing data collected at the panel user terminal, it is possible to collect and use data that is not limited to specific store data. It is also possible to use action data before and after action.

  In addition, by collecting and integrating information in stages, it is possible to create data that complements the portion where information shortage occurs on each page. Specifically, by creating summary information together with product information, it is possible to perform marketing analysis that complements the portion where information shortages occur by using summary information suitable for reading the panel user's behavior intentions. is there.

[Configuration of Purchasing Behavior Analysis System According to Embodiment 2]
A purchasing behavior analysis system according to the second embodiment will be described. FIG. 3 is a block diagram illustrating an example of the configuration of the purchase behavior analysis system according to the second embodiment. In FIG. 3, 200 indicates a panel user terminal, 300 indicates a purchasing behavior analysis server corresponding to the purchasing behavior analysis apparatus 100 in FIG. 1, 401 indicates an EC site, and 402 indicates the Internet. In the following, the description of the same points as in the first embodiment will be omitted or briefly described.

  In the example shown in FIG. 3, the case where the panel user terminal 200 and the purchase behavior analysis server 300 are connected via the Internet 402 is shown, but the present invention is not limited to this. For example, the panel user terminal 200 and the purchase behavior analysis server 300 may be connected using a dedicated line or may be connected by an arbitrary method.

[Panel user terminal]
Panel user terminal 200 is connected to EC site 401 and purchasing behavior analysis server 300 via Internet 402. In the example illustrated in FIG. 3, the case where there are a plurality of panel user terminals 200 is illustrated as an example, but the present invention is not limited to this, and the number of panel user terminals 200 may be one. The panel user terminal 200 includes a web browsing data DB 201 and a web browsing data transmission unit 211. The web browsing data DB 201 stores data used for various processes by the web browsing data transmission unit 211. The Web browsing data DB 201 corresponds to, for example, a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, a hard disk, an optical disk, or the like. The web browsing data DB 201 stores web browsing data including web pages browsed by panel users using the panel user terminal 200.

  FIG. 4 is a diagram illustrating an example of information stored in the Web browsing data DB according to the second embodiment. In the example illustrated in FIG. 4, the Web browsing data DB 201 stores “panel user ID”, “browsing number”, “browsing date / time”, “URL”, and “HTML source” in association with each other. “Panel user ID” in FIG. 4 indicates an ID for identifying a panel user, and “Viewing number” in FIG. 4 indicates information for uniquely identifying a record stored in the Web browsing data DB 201. Further, “browsing date / time” in FIG. 4 indicates the date / time viewed by the panel user, “URL” in FIG. 4 indicates the URL of the Web page browsed by the panel user, and “HTML source” in FIG. The HTML source of the web page browsed by the panel user is shown.

  In the example shown in FIG. 4, the Web browsing data DB 201 includes a panel user ID “00000001”, a browsing number “00000001”, a browsing date and time “2009-12-25 17:01:23”, and a URL “http: //www1.example. .com / foo.html ”and the HTML source“ <html> <head> ˜ </ head> <body> ˜ </ body> </ html> ”are stored in association with each other. That is, in the Web browsing data DB 201, the panel user identified by the panel user ID “00000001” as the record identified by the browsing number “00000001” is “2009-12-25 17:01 : 23 ”stores that the Web page at“ http://www1.example.com/foo.html ”has been browsed. In the Web browsing data DB 201, the HTML source of the Web page browsed by the panel user ID “00000001” is “<html> <head> ˜ </ head> <body> ˜ </ body> </ html>”. Remember that there is.

  In the example shown in FIG. 4, for convenience of description, <head> indicating the position of the header in the HTML source and <body> indicating the position of the content are described for the HTML source. However, actually, the Web browsing data DB 201 also stores the specific description content of the header and the specific description content of the body as an HTML source.

  The data stored in the web browsing data DB 201 is stored by the panel user terminal 200 every time a panel user browses a web page using the panel user terminal 200, for example.

  The Web browsing data transmission unit 211 has an internal memory that stores a program that defines various processing procedures and the like, and controls various processes. The web browsing data transmission unit 211 corresponds to an electronic circuit such as an application specific integrated circuit (ASIC), a field programmable gate array (FPGA), a central processing unit (CPU), and a micro processing unit (MPU).

  The web browsing data transmission unit 211 transmits the web browsing data stored in the web browsing data DB 201 to the purchase behavior analysis server 300 at a predetermined opportunity. The predetermined trigger is, for example, a timing when a certain amount of Web browsing data is newly accumulated in the Web browsing data DB 201, a regular interval such as every 5 minutes, every hour, or a timing when new Web browsing data is stored. The timing when the web browsing software starts, the timing when it ends, and the like are applicable. However, the present invention is not limited to this, and may be any timing.

  For example, the web browsing data transmission unit 211 creates an archive of web browsing data, and transmits the created archive to the purchase behavior analysis server 300. However, the present invention is not limited to this, and the Web browsing data transmission unit 211 may transmit the Web browsing data as it is without using the archive.

[Purchasing behavior analysis server]
The purchase behavior analysis server 300 will be described. As shown in FIG. 3, the purchase behavior analysis server 300 includes a text extraction rule DB 301, a dictionary data DB 302, a Web browsing data DB 303, a browsing text data DB 304, a browsing product data DB 305, a purchased product data DB 306, an action It has a summary data DB 307 and a marketing data DB 308. The purchase behavior analysis server 300 includes a Web browsing data receiving unit 311, a text extracting unit 312, a product information extracting unit 313, a summary generating unit 314, and a statistical analyzing unit 315.

  Here, the text extraction rule DB 301, the dictionary data DB 302, the Web browsing data DB 303, the browsing text data DB 304, the browsing product data DB 305, the purchased product data DB 306, the behavior summary data DB 307, and the marketing data DB 308 are: For example, a semiconductor memory device such as a RAM or a flash memory, a hard disk, an optical disk, or the like is applicable. The web browsing data receiving unit 311, the text extracting unit 312, the product information extracting unit 313, the summary generating unit 314, and the statistical analyzing unit 315 are, for example, electronic circuits such as ASIC, FPGA, CPU, and MPU. Applicable. In the example illustrated in FIG. 3, the text extraction rule DB 301 to the marketing data DB 308 are described as separate databases, and the web browsing data reception unit 311 to the statistical analysis unit 315 are described as separate processing units. However, the present invention is not limited to this, and may be a database in which some or all of the text extraction rule DB 301 to the marketing data DB 308 are integrated, and one of the web browsing data reception unit 311 to the statistical analysis unit 315. It is good also as a process part which integrated the part or all.

  With reference to FIG. 5, an overview of processing by the purchase behavior analysis server 300 in the second embodiment will be described. FIG. 5 is a diagram illustrating an overview of processing by the purchase behavior analysis server in the second embodiment. In FIG. 5, the same reference numerals are assigned to the databases and processing units corresponding to those in FIG. In the description with reference to FIG. 5, detailed description of the processing is omitted.

  As shown in (1) of FIG. 5, in the purchase behavior analysis server 300, the web browsing data receiving unit 311 receives the web browsing data collected by the panel user terminal 200 and stores it in the web browsing data DB 303. That is, the web browsing data receiving unit 311 collects the contents of the web pages browsed by the panel user and stores them in the web browsing data DB 303.

  Then, as shown in (2) of FIG. 5, the text extraction unit 312 extracts page information from the Web browsing data stored in the Web browsing data DB 303 using the text extraction rules stored in the text extraction rule DB 301. And stored in the browsing text data DB 304.

  Then, as shown in (3) of FIG. 5, the product information extraction unit 313 uses the dictionary data stored in the dictionary data DB 302 as the product information from the page information stored in the browse text data DB 304 as a panel. Product information about the product viewed by the user is extracted, and product information about the product purchased by the panel user is extracted. Then, the product information extraction unit 313 stores product information about the browsed product in the browse product data DB 305, and stores product information about the purchased product in the purchase product data DB 306.

  Further, as shown in (4) of FIG. 5, the summary generation unit 314 uses the dictionary data stored in the dictionary data DB 302 to generate summary information of the page information stored in the browse text data DB 304, and Stored in the summary data DB 307. For example, the summary generation unit 314 generates the summary information by dividing the page information for each cluster by an arbitrary method and assigning a representative keyword included in the page information.

  Then, as shown in (5) of FIG. 5, the statistical analysis unit 315 executes predetermined statistical analysis processing on the information stored in the browsing product data DB 305, the purchased product data DB 306, and the behavior summary data DB 307, Marketing data is generated and stored in the marketing data DB 308. For example, the statistical analysis unit 315 generates marketing data by performing statistical processing and visualization processing as the predetermined statistical analysis processing. Thereafter, the marketing data stored in the marketing data DB 308 is output as, for example, a marketing report.

  Returning to the description of FIG. The text extraction rule DB 301 is connected to the text extraction unit 312. The text extraction rule DB 301 stores rules for extracting page information from Web browsing data.

  FIG. 6 is a diagram illustrating an example of information stored in the text extraction rule DB according to the second embodiment. In the example illustrated in FIG. 6, the text extraction rule DB 301 stores a text extraction rule record including “URL condition”, “title condition”, “purchasing site”, “page type”, “Xpath condition”, and “data type”. Remember.

  Here, “URL condition” and “title condition” indicate a rule for extracting a Web page related to a product from a Web page browsed by a user included in the Web browsing data. Specifically, the URL condition indicates a URL of a Web page to be extracted as page information from Web pages browsed by a user included in the Web browsing data. The title condition indicates the title of a Web page to be extracted as page information among Web pages browsed by the user included in the Web browsing data.

  “Purchase site” indicates the name of the EC site when the Web page extracted by the “URL condition” and the “title condition” is the Web page of the EC site. The “page type” indicates whether the Web page extracted by the “URL condition”, the “title condition”, and the “purchase site” is a Web page for browsing information about a product. In the example illustrated in FIG. 6, the page type “product details” indicates a Web page for browsing information about the product, and the page type “purchase procedure” indicates a Web page for purchasing the product. It was.

  The “Xpath condition” is information indicating a portion to be extracted as page information from the extracted Web page, and the “data type” is a type of information extracted by the “Xpath condition”. The combination of items shown in FIG. 6 is an example, and the present invention is not limited to this, and any combination may be used.

  In the example shown in FIG. 6, the text extraction rule DB 301 includes a URL condition “http://foo.examlple.com/bbb.html”, a title condition “NULL”, a purchase site “foo”, and a page type “product details”. Records including the Xpath condition “// b / td [2] / span [3]” and the data type “sales name notation” are stored. That is, the text extraction rule DB 301 extracts the web page of the EC site “foo” that matches “http://foo.examlple.com/bbb.html” from the web browsing data, and among the extracted web pages, It stores that data indicating “sales name notation” is extracted from the location specified by “// b / td [2] / span [3]”. “NULL” indicates a case where no data is included.

  Returning to the description of FIG. The dictionary data DB 302 is connected to the product information extraction unit 313 and the summary generation unit 314. The dictionary data DB 302 is used when a keyword is extracted from the text included in the Web page or when the text is decomposed into morphemes. The dictionary data DB 302 is created by an existing arbitrary method. The dictionary data DB 302 is updated from time to time based on data collected from, for example, article text of specialized journals, EC site search queries, category information, and the like. The dictionary data DB 302 may be updated using an arbitrary method. The administrator of the purchase behavior analysis server 300 may update the dictionary data DB 302 manually. From the article text of a specialized journal, an EC site search query, category information, and the like. When the collected data is received, it may be automatically updated.

  The web browsing data DB 303 stores web browsing data. For example, each web browsing data transmitted by the panel user terminal 200 and received by the web browsing data receiving unit 311 is stored. The information stored in the Web browsing data DB 303 is the same as the information stored in the Web browsing data DB 201 and will not be described.

  The browsing text data DB 304 stores page information extracted from the Web browsing data by the text extraction unit 312. That is, the browsing text data DB 304 stores page information that is information included in a Web page related to a product among Web pages browsed by a user included in Web browsing data.

  FIG. 7 is a diagram illustrating an example of information stored in the browse text data DB according to the second embodiment. In the example shown in FIG. 7, the browsing text data DB 304 includes “viewing number”, “user ID”, “viewing date / time”, “site name”, “page type”, “text”, “title”, and “sales name notation”. ”,“ Price notation ”,“ product description ”,“ search keyword ”,“ product image ”,“ shop name ”,“ number notation ”,“ category ”, and“ brand ”are stored. However, the combination of items shown in FIG. 7 is an example, and the present invention is not limited to this, and arbitrary items may be used.

  Here, the “user ID” is information for identifying a user, and is included in, for example, Web browsing data. “Text” indicates the text included in the Web page. “Sales name notation” indicates the name of the product included in the Web page. “Product description text” indicates a text description about the product among the text included in the Web page. “Search keyword” indicates a keyword set for the Web page. For example, the search keyword is set in the tag of the Web page and set as meta information. “Product image” indicates an image of a product among images included in the Web page. “Shop name” indicates the name of a store that describes a product on the Web page or the name of a store that sells the product on the Web page. “Number notation” indicates the number of products purchased by the user when the Web page indicates that the user has purchased the product. “Category” indicates a category of a product included in the Web page. “Brand” indicates a brand assigned to a product.

  Here, the “browsing number” is the same number as the “browsing number” shown in FIG. In addition, “site name”, “page type”, “text”, “title”, “sales name notation”, “price notation”, “product description”, “search keyword”, “product image”, and “shop name” "," Number notation "," category ", and" brand "are extracted from the Web browsing data by the product information extraction unit 313 based on the dictionary data DB 302, respectively. For example, by setting “site name” to “brand” as “data type” in the text extraction rule, “site name” to “brand” are extracted from the Web browsing data based on the text extraction rule. Further, “sales name notation”, “price notation”, “shop name”, “number notation”, “category”, and “brand” are extracted from the web browsing data by extracting keywords from the text based on the dictionary data DB 302. Is done.

  In the example shown in FIG. 7, the browsing text data DB 304 includes a browsing number “00000001”, a user ID “00000001”, a browsing date and time “2009-12-25 17:01:23”, a site name “Site A”, and a page type “ "Product details", the text "This product is um ...", the title "product xx", the sales name notation "product xx", the price notation "1200 yen", the product description "this product is ar ..." and the search keyword " A record including “NULL”, product image “(image data)”, shop name “shop a”, number notation “NULL”, category “T-shirt”, and brand “brand X” is stored. In the example illustrated in FIG. 7, for convenience of description, “(image data)” is stored as a product image. However, “(image data)” indicates image data.

  That is, in the example illustrated in FIG. 7, the browsing text data DB 304 is identified by the browsing number “00000001” among the Web browsing data browsed by the user “00000001” on “2009-12-25 17:01:23”. Stores page information about the record. Further, the browsing text data DB 304 is a Web page for browsing the details of the product with the site name of the Web page being “Site A”, and the Web page includes the text “This product is pun ...”. Then, the fact that the title of the Web page is “product” is stored. In addition, the browsing text data DB 304 stores that the information about the product “product XX” is included in the Web page, the price is “1200 yen”, and there is an explanatory note “this product is pun ...”. . In addition, an image “(image data)” is included in the Web page, the store name relating to the Web page is “Shop a”, the product category is “T-shirt”, and the product brand is “Brand X”. Remember that.

  Returning to the description of FIG. The browsing product data DB 305 stores product information extracted from a Web page for browsing information about products. Specifically, the browsing product data DB 305 stores product information included in the record of the browsing text data DB 304 whose page type is “product details”.

  FIG. 8 is a diagram illustrating an example of information stored in the browsing product data DB according to the second embodiment. In the example illustrated in FIG. 8, the browsing product data DB 305 includes “viewing number”, “site name”, “shop name”, “sales name notation”, “price notation”, “category”, “brand”, and “product image”. ”Is stored. Note that the combination of items shown in FIG. 8 is an example, and the present invention is not limited to this, and may be arbitrary.

  For example, the browsing product data DB 305 includes a browsing number “00000001”, a site name “site A”, a shop name “shop a”, a sales name notation “product xx”, a price notation “1200 yen”, and a category “T-shirt”. The record of the browsing product including the brand “brand X” and the product image “(image data)” is stored.

  Returning to the description of FIG. 3, the purchased product data DB 306 stores product information extracted from a Web page for purchasing a product. Specifically, the purchased product data DB 306 stores product information included in a record of the browse text data DB 304 whose page type is “purchase procedure”.

  FIG. 9 is a diagram illustrating an example of information stored in the purchased product data DB according to the second embodiment. In the example illustrated in FIG. 9, the purchased product data DB 306 includes “view number”, “site name”, “shop name”, “sales name notation”, “price notation”, “number notation”, “category”, and “brand”. ”And“ product image ”are stored. The combination of items shown in FIG. 9 is an example, and the present invention is not limited to this, and any combination of items may be used. 8 and FIG. 9, the purchased product data DB 306 is product information about the Web page indicating the purchase procedure, and “number notation” is added to the item of the browse text data DB 304.

  For example, the purchase product data DB 306 includes a browsing number “00000002”, a site name “site A”, a shop name “shop a”, a sales name notation “product xx”, a price notation “1200 yen”, and a number notation “1”. Records of purchased products including category “T-shirt”, brand “brand X”, and product image “(image data)” are stored.

  Returning to the description of FIG. The behavior summary data DB 307 stores summary information that is a summary of page information. FIG. 10 is a diagram illustrating an example of information stored in the behavior summary data DB according to the second embodiment. In the example illustrated in FIG. 10, the behavior summary data DB 307 stores a record including “user ID”, “viewing number”, “cluster number”, and “representative keyword”. The combination of items shown in FIG. 10 is an example, and the present invention is not limited to this, and any combination of items may be used.

  Here, the “cluster number” will be briefly described. A later-described summary generation unit 314 classifies records included in the browse text data DB 304 into a plurality of groups. The cluster number is information for identifying each group. The representative keyword indicates a representative keyword included in each group.

  In the example illustrated in FIG. 10, the behavior summary data DB 307 stores a behavior summary record including the user ID “00000001”, the browsing number “00000001”, the cluster number “00000001”, and the representative keyword “T-shirt, brand X”. . That is, the behavior summary data DB 307 stores that the behavior summary record identified by the user browsing number “00000001” of “00000001” is classified into the group identified by the cluster number “00000001”. It is stored that the representative keyword of the group identified by the number “00000001” is “T-shirt, brand X”.

  The marketing data DB 308 stores marketing data that is a processing result by the statistical analysis unit 315. 11 and 12 are diagrams illustrating an example of marketing data stored by the marketing data DB in the second embodiment. FIG. 11 shows marketing data when brands are ranked for each category of products within an arbitrary period. Moreover, FIG. 12 shows marketing data when the horizontal axis is an item and the vertical axis is an average amount of items for each category. Data stored in the marketing data DB 308 is output as a marketing report.

  Returning to the description of FIG. The web browsing data receiving unit 311 acquires web browsing data by receiving it from the panel user terminal 200. For example, when receiving the web browsing data archive, the web browsing data receiving unit 311 decompresses the received web browsing data archive and stores the web browsing data obtained as a result of the decompression in the web browsing data DB 303.

  For example, when receiving a text extraction rule from the administrator of the purchase behavior analysis server 300, the text extraction unit 312 stores the text extraction rule 312 in the text extraction rule DB 301. For example, the administrator arbitrarily executes registration and change of the text extraction rule. For example, at the registration timing, the text extraction unit 312 registers a text extraction rule by the administrator, and stores the registered text extraction rule in the text extraction rule DB 301.

  Note that the text extraction unit 312 may detect a change in the site structure of the EC site by monitoring page information stored in the browse text data DB 304 and change the text extraction rule based on the detection result.

  In addition, the text extraction unit 312 executes an extraction process on the Web browsing data acquired by the Web browsing data receiving unit 311 according to a predetermined extraction condition, so that the Web page related to the product among the Web pages browsed by the user is displayed. The page information that is included is extracted. Specifically, the text extraction unit 312 reads out new Web browsing data from the browsing text data DB 304 when the new Web browsing data is in the browsing text data DB 304, and reads out the new Web browsing data based on the text extraction rules. A Web page for purchasing a product is specified from a Web page included in the Web browsing data, and a Web page for browsing information about the product is specified. Then, the text extracting unit 312 extracts page information that is information included in the identified Web page based on the main extraction rule. For example, the text extraction unit 312 extracts page information by extracting a part specified by the text extraction rule from an HTML source included in the Web browsing data.

  For example, the text extraction unit 312 executes an extraction process at an arbitrary timing. The arbitrary timing corresponds to, for example, a timing at which a certain amount of new Web browsing data is accumulated when it is executed at regular intervals such as every 5 minutes or every hour. An example of the flow of processing by the text extraction unit 312 will be described later using a flowchart.

  The product information extraction unit 313 extracts product information that is information about the product included in the page information from the page information extracted by the text extraction unit 312. Specifically, the product information extraction unit 313 extracts the first product information as product information from the Web page for purchasing the product, and the product information included in the Web page for browsing information about the product. Certain second product information is extracted.

  For example, the product information extraction unit 313 acquires one record from the browsing text data DB 304 when new page information is in the browsing text data DB 304, and if the page type of the acquired record is “product details”, Product information is stored in the browsed product data DB 305. The product information extraction unit 313 stores product information in the purchased product data DB 306 if the page type of the acquired record is “purchase procedure”. Explaining with a more detailed example, the product information extraction unit 313 stores the data of items corresponding to product information among the items included in the record acquired from the browse text data DB 304 in the browse product data DB 305 or the purchased product. Store in the data DB 306. An example of the flow of processing by the merchandise information extraction unit 313 will be described later using a flowchart.

  The product information extraction unit 313 extracts keywords from the body of the page information included in the record based on the dictionary data DB 302, extracts keywords corresponding to the product information from the extracted keywords, and browses the product data DB 305 or purchased products. You may store in data DB306.

  The summary generation unit 314 generates summary information that is a summary of information included in the page information from the page information extracted by the text extraction unit 312. Specifically, the summary generation unit 314 acquires a page information record from the browse text data DB 304 and performs cluster processing on the acquired page information record. In other words, based on the similarity between the page information records, a plurality of page information records for the user to be processed are classified into groups. Then, the summary generation unit 314 performs keyword extraction processing. That is, the summary generation unit 314 extracts a keyword representing a record included in the cluster for each cluster. Then, the summary generation unit 314 performs a summary generation process. That is, the summary generation unit 314 generates a summary by associating the keyword for each cluster with the cluster number. Then, the summary generation unit 314 stores the generated summary in the behavior summary data DB 307. An example of processing by the summary generation unit 314 will be described later using a flowchart.

  The statistical analysis unit 315 performs a predetermined statistical analysis process on the product information extracted by the product information extraction unit 313 and the summary information generated by the summary generation unit 314. Specifically, the statistical analysis unit 315 acquires the product information stored in the browsed product data DB 305, acquires the product information stored in the purchased product data DB 306, and uses the summary information stored in the behavior summary data DB 307. get. Then, the statistical analysis unit 315 performs a statistical analysis process on the acquired data, and stores marketing data as a processing result in the marketing data DB 308. The predetermined statistical analysis process used by the statistical analysis unit 315 may be an arbitrary statistical analysis process. An example of processing by the statistical analysis unit 315 will be described later using a flowchart.

[Processing by panel user terminal]
FIG. 13 shows an example of the flow of processing by the panel user terminal 200 in the second embodiment. FIG. 13 is a flowchart illustrating an example of a process flow of the panel user terminal according to the second embodiment.

  As shown in FIG. 13, in the panel user terminal 200, when there is new web browsing data (Yes in step S201), the web browsing data transmitting unit 211 creates an archive of new web browsing data (step S202). Then, the web browsing data transmission unit 211 transmits an archive of the created web browsing data to the purchase behavior analysis server 300 (step S203).

  In the example illustrated in FIG. 13, the Web browsing data transmission unit 211 creates and transmits an archive, but the present invention is not limited to this. For example, the Web browsing data transmission unit 211 may transmit the Web browsing data itself without creating an archive.

[Processing by purchasing behavior analysis server]
An example of the flow of processing by the purchase behavior analysis server 300 is shown. In the following, an example of the flow of processing by the Web browsing data receiving unit 311, an example of the flow of processing by the text extraction unit 312, an example of the flow of processing by the product information extraction unit 313, and the flow of processing by the summary generation unit 314 And an example of the flow of processing by the statistical analysis unit 315 will be described in order.

[Processing by Web browsing data receiver]
An example of the flow of processing by the Web browsing data receiving unit 311 in the second embodiment will be described with reference to FIG. FIG. 14 is a flowchart illustrating an example of a processing flow by the Web browsing data receiving unit according to the second embodiment.

  As illustrated in FIG. 14, when the web browsing data receiving unit 311 receives an archive of web browsing data (Yes in step S301), the web browsing data receiving unit 311 decompresses the received web browsing data archive (step S302). Then, the web browsing data receiving unit 311 stores the web browsing data obtained as a result of the decompression in the web browsing data DB 303 (step S303).

  In the example shown in FIG. 14, the Web browsing data receiving unit 311 has shown a case where the archive is decompressed and stored in the Web browsing data DB 303, but the present invention is not limited to this. For example, the Web browsing data receiving unit 311 may store the archive as it is without decompressing the archive.

  In the example illustrated in FIG. 14, the panel user terminal 200 transmits an archive of Web browsing data. However, the present invention is not limited to this, and when the panel user terminal 200 transmits the web browsing data itself, the web browsing data receiving unit 311 does not perform the decompression process and receives the received web browsing. Store the data.

[Registration process by text extractor]
An example of the flow of registration processing by the text extraction unit 312 in the reason 2 will be described with reference to FIG. FIG. 15 is a flowchart illustrating an example of a registration process performed by the text extracting unit according to the second embodiment.

  As shown in FIG. 15, the text extraction unit 312 registers the text extraction rule by the administrator (step S402) at the registration timing (Yes in step S401), and stores the registered text extraction rule in the text extraction rule DB 301. (Step S403). For example, the text extraction unit 312 receives the text extraction rule from the administrator, registers the text extraction rule, and stores the received text extraction rule in the text extraction rule DB 301.

[Extraction process by headquarters extraction unit]
An example of the flow of extraction processing by the text extraction unit 312 according to the second embodiment will be described with reference to FIG. FIG. 16 is a flowchart illustrating an example of the flow of extraction processing by the text extraction unit in the second embodiment.

  As illustrated in FIG. 16, the text extraction unit 312 determines whether new web browsing data exists in the web browsing data DB 303 (step S501). If the text extraction unit 312 determines that there is one (Yes at Step S501), it acquires one page information record from the Web browsing data DB 303 (Step S502). Then, the text extraction unit 312 determines whether or not a text extraction rule has been registered in the text extraction rule DB 301 (step S503). If it is determined (Yes in step S503), the text extraction unit 312 determines the text from the HTML source included in the web browsing data. A part specified by the extraction rule is extracted (step S504). That is, the text extracting unit 312 extracts page information by executing an extraction process on the Web browsing data acquired by the Web browsing data receiving unit 311 according to a predetermined extraction condition. Then, the text extracting unit 312 stores the page information in the browse text data DB (step S505).

  Then, the text extracting unit 312 returns to step S501 and repeats the processes from step S501 to S505 until it determines that there is no new web browsing data in the web browsing data DB 303. If the text extraction unit 312 determines that there is no text extraction rule in step S503 described above (No in step S503), the text extraction unit 312 ends the process. If the text extraction unit 312 determines in step S501 that there is no new web browsing data in the web browsing data DB 303 (No in step S501), the text extraction unit 312 ends the process.

  In addition, said process procedure is not limited to said order, You may change suitably in the range which does not contradict a process content. For example, first, step S501 and step S503 in FIG. 16 are executed. At this time, when it is determined that there is new Web browsing data and there is a text extraction rule, step S502 in FIG. Step S504 and step S505 may be executed.

[Processing by product information extraction unit]
FIG. 17 shows an example of the flow of processing performed by the product information extraction unit 313 in the second embodiment. FIG. 17 is a flowchart illustrating an example of a process flow by the product information extraction unit according to the second embodiment.

  As illustrated in FIG. 17, when new page information is present in the browse text data DB 304 (Yes in step S601), the product information extraction unit 313 acquires one record from the browse text data DB 304 (step S602). For example, the product information extraction unit 313 determines whether there is a new record that has not been processed in the browse text data DB 304 or acquires a record each time a new record is stored in a stored procedure.

  Then, the product information extraction unit 313 extracts keywords from the body of the page information included in the record based on the dictionary data DB 302 (step S603). That is, the product information extraction unit 313 extracts a keyword that matches a phrase stored in the dictionary data DB 302 that is updated from time to time based on data collected from article text of specialized journals, EC site search queries, category information, and the like. To do.

  Then, the product information extraction unit 313 determines whether the page type is “product details” or “purchase procedure” (step S604). If the page type is “purchase procedure” (step S604 purchase procedure), the product information Is stored in the purchased product data DB 306 (step S605). On the other hand, if the page type is “product details” (step S604 product details), the product information extraction unit 313 stores the product information in the browsed product data DB 305 (step S606).

  For example, the product information extraction unit 313 stores data of items corresponding to product information among items included in the record acquired from the browse text data DB 304 in the browse product data DB 305 or the purchased product data DB 306. To explain with a more detailed example, the product information extraction unit 313 includes “view number”, “site name”, “shop name”, “sales name notation”, “price notation” among the items of the record. “Number notation”, “category”, “brand”, and “product image” are stored in the purchased product data DB 306. Similarly, the product information extraction unit 313 includes “view number”, “site name”, “shop name”, “sales name notation”, “price notation”, “category”, and “category” among the items of the record. The “brand” and the “product image” are stored in the browsing product data DB 305. Here, each item included in the record does not necessarily include data. In this case, the product information extraction unit 313 sets “NULL” without inputting any data for items that do not include data.

  And the merchandise information extraction part 313 returns to step S601, and repeats the process to the above-mentioned step S601-S606 until it determines with there being no new page information. On the other hand, when the product information extraction unit 313 determines in step S601 that there is no new page information (No in step S601), the product information extraction unit 313 ends the process. In addition, said process procedure is not limited to said order, You may change suitably in the range which does not contradict a process content. For example, step S603 described above may be omitted.

[Processing by summary generator]
An example of the flow of processing by the summary generation unit 314 according to the second embodiment will be described with reference to FIG. FIG. 18 is a flowchart illustrating an example of a processing flow by the summary generation unit according to the second embodiment.

  As illustrated in FIG. 18, when the processing target period and the processing target panel user ID are specified by the administrator (Yes in step S701), the summary generation unit 314 matches or corresponds to the specified period and the panel user ID. Is acquired from the browsing text data DB 304 (step S702). For example, when the processing target period “2009-12-25 00:00:00 to 2009-12-26 00:00:00” and the processing target panel user ID “00000001” are specified, the summary generation unit 314 From the browsing text data DB 304, a record is acquired that includes the browsing date and time included in “2009-12-25 00:00:00 to 2009-12-26 00:00:00” and the panel user ID “00000001”.

  And the summary production | generation part 314 decomposes | disassembles the text of each browsing page into a morpheme based on dictionary data DB (step S703). That is, the summary generation unit 314 decomposes the body of the page included in the acquired record into morphemes based on the dictionary data DB 302.

  Then, the summary generation unit 314 clusters browsing pages using the keywords decomposed into morphemes (step S704). That is, the summary generation unit 314 calculates the similarity between records using a keyword obtained for each record, and classifies a plurality of records for users to be processed into groups based on the calculated similarity between records. To do.

  Then, the summary generation unit 314 stores the browsing number and the cluster number in the behavior summary data DB 307 (step S705). That is, the summary generation unit 314 issues an arbitrary cluster number and assigns it to each group of records classified in step S704. Then, the summary generation unit 314 associates the cluster number with the browsing number of the record included in the group to which the cluster number is assigned, and stores it in the behavior summary data DB 307. For example, a case where a record including the browsing number “00000001” and a record including “00000002” are classified into the same group and the cluster number “00000001” is assigned will be described. In this case, the summary generation unit 314 stores the association between the browsing number “00000001” and the cluster number “00000001” in the behavior summary data DB 307, and associates the browsing number “00000002” with the cluster number “00000001”. Stored in the summary data DB 307. In addition, the panel user ID to be processed is also stored. At this stage, the representative keyword is blank among the items of the behavior summary data DB 307 shown in FIG.

  Then, the summary generation unit 314 sets the target cluster number to “00000001” (step S706), and acquires page information of the browsing number assigned with the target cluster number (step S707). Then, the summary generation unit 314 decomposes the body of each page information into morphemes based on the dictionary data DB 302 (step S708), and extracts representative keywords from the keywords decomposed into morphemes (step S709). For example, the summary generation unit 314 extracts the representative keyword “T-shirt, brand X” from the page information of the cluster number “00000001”.

  For example, the summary generation unit 314 may use any method as a method for extracting keywords. For example, TFIDF (term frequency inverse document frequency) or c-value may be used.

  Then, the summary generation unit 314 stores the representative keyword in the behavior summary data DB 307 (step S710). That is, the representative keyword “T-shirt, brand X” is stored in the record including the cluster number “00000001”.

  Then, the summary generation unit 314 increments the target cluster number (step S711). For example, the summary generation unit 314 sets the target cluster number as “00000002”. Then, the summary generation unit 314 determines whether there is a cluster with the target cluster number (step S712). For example, in step S704 described above, when the records are classified into four groups and cluster numbers “00000001” to “00000004” are assigned, the summary generation unit 314 determines that there is (Yes in step S712). Then, the process returns to step S707 described above, and the process is repeated. On the other hand, the summary generation unit 314 determines that there is no cluster number “00000001” to “00000004” and the incremented target cluster number is “00000005” (No in step S712). End the process.

  In addition, said process procedure is not limited to said order, You may change suitably in the range which does not contradict a process content. For example, in the example shown in FIG. 18, the case where a keyword is extracted from each record and clustering is performed based on the similarity of the keyword is shown. However, the present invention is not limited to this, and an arbitrary method is used. You can cluster. In the example shown in FIG. 18, the case where the morpheme is decomposed in step S703 and the morpheme is also decomposed in step S708 is shown, but the present invention is not limited to this. For example, in step S708, by using the processing result in step S703, it is not necessary to execute the process of decomposing into morphemes.

[Processing by statistical analysis unit]
An example of the flow of processing by the statistical analysis unit 315 according to the second embodiment will be described with reference to FIG. FIG. 19 is a flowchart illustrating an example of a process flow by the statistical analysis unit according to the second embodiment.

  As shown in FIG. 19, when the analysis target period, the analysis target user condition, and the analysis method are specified by the administrator (Yes in step S801), the statistical analysis unit 315 reads the browsing product data DB 305, the purchased product data DB 306, Records with matching periods and conditions are acquired from the behavior summary data DB 307 (step S802).

  Then, the statistical analysis unit 315 performs statistical analysis (step S803) and outputs a statistical analysis result (step S804). That is, the statistical analysis unit 315 performs a predetermined statistical analysis process on the product information extracted by the product information extraction unit 313 and the summary information generated by the summary generation unit 314, and the processing result is displayed in the marketing data DB 308. To store.

[Effect of Example 2]
As described above, according to the second embodiment, the purchase behavior analysis server 300 acquires Web browsing data including a Web page browsed by the user on the panel user terminal 200. Then, the purchase behavior analysis server 300 performs extraction processing on the acquired Web browsing data according to a predetermined extraction condition, and thereby page information that is information included in the Web page related to the product among the Web pages browsed by the user. To extract. Then, the purchasing behavior analysis server 300 extracts product information that is information about the product included in the page information from the extracted page information, and generates summary information that is a summary of the information included in the page information from the page information. To do. Then, the purchase behavior analysis server 300 performs a predetermined statistical analysis process on the product information and the summary information. As a result, according to the purchasing behavior analysis apparatus 100 according to the first embodiment, marketing analysis can be appropriately executed.

  Further, according to the second embodiment, the purchase behavior analysis server 300 specifies a Web page for purchasing a product as a Web page related to the product, specifies a Web page for browsing information about the product, and specifies the Web page. The page information that is information included in the Web page is extracted. Then, the purchase behavior analysis server 300 extracts the first product information that is product information included in the Web page for purchasing the product, and the product information included in the Web page for browsing information about the product. Certain second product information is extracted. Then, the purchase behavior analysis server 300 performs a predetermined analysis process on the summary information, the first product information, and the second product information. As a result, marketing analysis can be performed using information about the product viewed by the user and information about the product purchased by the user.

  That is, the purchase behavior analysis server 300 acquires information from a page that cannot be obtained by post-crawl, such as a purchase procedure page, by acquiring and using Web browsing data from the panel user terminal 200. Marketing analysis is possible and marketing analysis is possible with high accuracy.

  FIG. 20 is a diagram for illustrating the effect of the second embodiment. As shown in Fig. 20, there are multiple marketing analysis methods, but none of them can analyze behavior before and after purchase, cannot collect cross-sectional data, and cannot target a wide range of target products. There may be a time lag between the time when the purchase action is taken and the time when the data is collected, or there is a problem in accuracy. The “proposed method” in FIG. 20 indicates a processing method performed by the purchase behavior analysis server 300 according to the second embodiment.

  For example, as shown in the “questionnaire” in FIG. 20, in the method of collecting data using a questionnaire, the time lag between the time when the purchase action is taken and the time when the data is collected is large. There are many omissions and memory differences, resulting in low reliability. Further, as shown in “contact investigation by RFID” in FIG. 20, in the method of attaching an RFID tag to a product and using information on the product tag that the consumer has contacted, only the product contact with the target product is used. As a result, the behavior before and after purchasing cannot be analyzed. Further, in the method using the information about the tag of the product that the consumer has contacted, the RFID must be attached to the target product, so that a wide range of target products cannot be targeted. Further, as shown in “in-store lead analysis” in FIG. 20, the method of analyzing a consumer's lead and line of sight from a surveillance camera in the store is closed in the store, and sharing between stores is difficult. As a result, cross-sectional data collection is not possible. Moreover, in the method of analyzing a consumer's conductor and line of sight from a surveillance camera in the store, the target product is limited to a product handled in the store, and it may take time to interpret the behavior of the conductor. Further, as shown in “Web log analysis of EC site” in FIG. 20, the method of analyzing the Web log obtained at a specific EC site is closed within the EC site and is difficult to share between sites. , Cross-sectional data collection is not possible. Further, in the method of analyzing a Web log obtained at a specific EC site, the target product is limited to products handled at the EC site. Further, as shown in “POS data etc.” in FIG. 20, the marketing analysis method using POS data can only use the data about purchasing, cannot analyze the behavior before and after purchasing, and has the same series. As a result, it is difficult to collect data across stores, and it is difficult to share data between stores. Further, in the method of marketing analysis using POS data, the target products are limited to the registered products to which the JAN code is assigned.

  On the other hand, as shown in FIG. 20, in the processing method by the purchase behavior analysis server 300 in the second embodiment, as a result of using the web browsing data by the panel user terminal 200, both browsing data and purchasing data are used. It is possible to analyze the behavior before and after purchase. Further, in the processing method by the purchase behavior analysis server 300, the panel user terminal 200 analyzes using the web browsing data, so that it is possible to collect data transversely without being limited to a specific store. As a result, it is possible to analyze products and price comparison behavior across stores. Further, in the processing method by the purchase behavior analysis server 300, it is possible to use data on all products accessed by the panel user by using Web browsing data by the panel user terminal 200. Moreover, in the processing method by the purchase behavior analysis server 300, Web browsing data can be acquired and used from the panel user terminal 200 almost in real time, and between the time when the panel user takes the purchase behavior and the time when the data is collected. It is possible to suppress the time lag. In the processing method by the purchase behavior analysis server 300, the accuracy can be improved by improving the accuracy of text mining.

  In the third embodiment, a case where personal information is deleted from Web browsing data transmitted from the panel user terminal 200 will be described. Hereinafter, a case where the panel user terminal 200 transmits personal information from the Web browsing data to the purchase behavior analysis server 300 after deleting the personal information will be described. However, the present invention is not limited to this, and the purchase behavior analysis server 300 is used. However, the personal information may be deleted from the received Web browsing data. Further, both the panel user terminal 200 and the purchase behavior analysis server 300 may delete the personal information from the web browsing data.

  FIG. 21 is a block diagram illustrating an example of the configuration of the purchasing behavior analysis system according to the third embodiment. As shown in FIG. 21, in the third embodiment, the panel user terminal 200 further includes a personal information data DB 501 and a personal information removal processing unit 502. In FIG. 21, the same reference numerals as those in FIG. 3 are assigned to the processing units and databases similar to those in FIG.

  The personal information data DB 501 stores personal information of panel users. FIG. 22 is a diagram illustrating an example of information stored in the personal information data DB according to the third embodiment. In the example illustrated in FIG. 22, the personal information data DB 501 stores “name”, “phone number”, “address”, “card number”, and the like. However, the present invention is not limited to this, and the personal information data DB 501 may store data for arbitrary items.

  For example, the personal information data DB 501 stores a name “Taro Teshinshin”, a telephone number “012-345-6789”, an address “Musashino City XXX Tokyo”, and a card number “0123 4567 8901 2345”.

  When the web browsing data is acquired from the web browsing data DB 201, the personal information removal processing unit 502 deletes data that matches or corresponds to the personal information stored in the personal information data DB 501 from the acquired web browsing data. The personal information removal processing unit 502 may replace the personal information with an arbitrary character string when deleting the personal information from the Web browsing data.

  Here, processing by the panel user terminal 200 according to the third embodiment will be described with reference to FIG. FIG. 23 is a flowchart illustrating processing by the panel user terminal according to the third embodiment.

  As shown in FIG. 23, in the panel user terminal 200, when there is new Web browsing data (Yes in step S901), the personal information removal processing unit 502 adds text that matches the personal information data in the new Web browsing data. It is determined whether or not there is (step S902). If the personal information removal processing unit 502 determines that there is one (Yes at Step S902), the text matching the personal information data is replaced with an arbitrary character (Step S903). That is, the personal information removal processing unit 502 deletes personal information data from the Web browsing data.

  Then, the web browsing data transmission unit 211 creates an archive of the web browsing data after the personal information is deleted (step S904), and transmits the created web browsing data archive to the purchase behavior analysis server 300 (step S905). ).

  In the example shown in FIG. 23, the Web browsing data transmission unit 211 creates and transmits an archive, but the present invention is not limited to this. For example, the Web browsing data transmission unit 211 may transmit the Web browsing data itself without creating an archive.

  In the above description, the case where the personal information data DB 501 matches or corresponds to the case of deleting from the Web browsing data has been described. However, the present invention is not limited to this, and the personal information is obtained by an arbitrary method. You can detect and delete it.

[Effect of Example 3]
As described above, according to the third embodiment, the Web browsing data from which the personal information has been deleted is used. Therefore, after the personal information is removed, the purchase behavior analysis server 300 holds the data. As a result, it is possible to provide a secure service in consideration of personal information.

  Although the embodiments of the present invention have been described so far, the present invention may be implemented in other embodiments besides the above-described embodiments. Therefore, other embodiments will be described below.

[System configuration]
Also, among the processes described in this embodiment, all or part of the processes described as being performed automatically can be performed manually, or the processes described as being performed manually can be performed. All or a part can be automatically performed by a known method. In addition, the processing procedures, control procedures, specific names, and information including various data and parameters shown in the above-mentioned document and drawings (FIGS. 1 to 23) are arbitrarily changed unless otherwise specified. be able to.

  Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. That is, the specific form of distribution / integration of each device is not limited to the one shown in the figure, and all or a part of the distribution / integration is functionally or physically distributed in arbitrary units according to various loads and usage conditions. Can be integrated and configured.

  For example, FIG. 24 is used to show an example of a variation of the configuration of the purchase behavior analysis system. FIG. 24 is a diagram illustrating an example of a variation of the configuration of the purchase behavior analysis system. As shown in FIG. 24, the statistical analysis unit 601 and the marketing data DB 602 may be separated from the purchase behavior analysis server 300 as another external device 600 and connected via a network. The statistical analysis unit 601 and the marketing data DB 602 correspond to the statistical analysis unit 315 and the marketing data DB 308 in FIG. 3, respectively. In this case, the external device 600 accesses the browsing product data DB 603, the purchased product data DB 604, and the behavior summary data DB 605 of the purchase behavior analysis server 300 to acquire data, and executes statistical analysis processing on the acquired data. Create marketing data with. In the example shown in FIG. 24, when there are a plurality of external devices, the types of accessible data may be limited for each external device. As a result, it is possible to realize a business in which a company that operates a purchasing behavior analysis system sells access fees to analysis devices and analysis source data to external marketers.

[program]
FIG. 25 is a diagram illustrating that information processing by the purchase behavior analysis program is specifically realized using a computer. As illustrated in FIG. 25, the computer 3000 includes, for example, a memory 3010, a CPU (Central Processing Unit) 3020, a hard disk drive interface 3030, a disk drive interface 3040, a serial port interface 3050, a video adapter 3060, Network interface 3070. Each part of the computer 3000 is connected by a bus 3100.

  Note that the configuration of the computer illustrated in FIG. 25 is an example, and the configuration of the computer is not limited thereto. For example, in the example shown in FIG. 25, the computer 3000 has the hard disk drive 3080 and the hard disk drive 3080 stores the program data 3084 and the like, but the present invention is not limited to this. For example, the computer 3000 may not have the hard disk drive 3080. In this case, for example, the computer 3000 may store the program data 3084 or the like in the memory 310, and may store the program data 3084 or the like in an arbitrary storage area. The computer 3000 may be an arbitrary information processing apparatus, and may be an information processing apparatus such as a mobile phone, a PHS (Personal Handyphone System) terminal, a mobile communication terminal, or a PDA (Personal Digital Assistant). .

  The memory 3010 includes a ROM 3011 and a RAM 3012 as illustrated in FIG. The ROM 3011 stores a boot program such as BIOS (Basic Input Output System). The hard disk drive interface 3030 is connected to the hard disk drive 3080 as illustrated in FIG. The disk drive interface 3040 is connected to the disk drive 3090 as illustrated in FIG. For example, a removable storage medium such as a magnetic disk or an optical disk is inserted into the disk drive 3090. The serial port interface 3050 is connected to, for example, a mouse 3051 and a keyboard 3052 as illustrated in FIG. The video adapter 3060 is connected to a display 3061, for example, as illustrated in FIG.

  Here, as illustrated in FIG. 25, the hard disk drive 3080 stores, for example, an OS 3081, an application program 3082, a program module 3083, and program data 3084. In other words, the update program according to the disclosed technology is stored in, for example, the hard disk drive 3080 as the program module 3083 in which instructions executed by the computer are described. Specifically, a program module in which each of the procedures for executing the same information processing as the Web browsing data transmitting unit 211 and the Web browsing data receiving unit 311 to the statistical analysis unit 315 described in the above embodiment is described in the hard disk drive 3080. Is remembered.

  Further, data used for information processing by the purchase behavior analysis program, such as data stored in the Web browsing data DB 201 and the text extraction rule DB 301 to the marketing data DB 308 described in the above-described embodiment, is set as program data 3084, for example, on a hard disk Stored in drive 3080. The CPU 3020 reads the program module 3083 and program data 3084 stored in the hard disk drive 3080 to the RAM 3012 as necessary, and executes various procedures.

  Note that the program module 3083 and the program data 3084 related to the purchase behavior analysis program are not limited to being stored in the hard disk drive 3080. For example, the program module 3083 and the program data 3084 may be stored in a removable storage medium. In this case, the CPU 3020 reads data via a removable storage medium such as a disk drive. Similarly, the program module 3083 and program data 3084 related to the update program may be stored in another computer connected via a network (LAN (Local Area Network), WAN (Wide Area Network), etc.). . In this case, the CPU 3020 reads various data by accessing another computer via the network interface.

[Others]
Note that the purchase behavior analysis program described in the present embodiment can be distributed via a network such as the Internet. Further, the purchase behavior analysis program can be executed by being recorded on a computer-readable recording medium such as a hard disk, a flexible disk (FD), a CD-ROM, an MO, and a DVD, and being read from the recording medium by the computer. .

DESCRIPTION OF SYMBOLS 100 Purchasing behavior analysis apparatus 101 Acquisition part 102 Page information extraction part 103 Product information extraction part 104 Summary generation part 105 Statistical analysis part 200 Panel user terminal 201 Web browsing data DB
211 Web browsing data transmission unit 300 Purchasing behavior analysis server 301 Text extraction rule DB
302 Dictionary data DB
303 Web browsing data DB
304 Browsing text data DB
305 Browsing product data DB
306 Purchased Product Data DB
307 Action Summary Data DB
308 Marketing Data DB
311 Web browsing data reception unit 312 Text extraction unit 313 Product information extraction unit 314 Summary generation unit 315 Statistical analysis unit 401 EC site 402 Internet 501 Personal information data DB
502 Personal information removal processing unit 600 External device

Claims (9)

An acquisition unit for acquiring web browsing data including a web page browsed by a user at a user terminal;
By performing an extraction process on the Web browsing data acquired by the acquisition unit according to a predetermined extraction condition, page information that is information included in a Web page related to a product is extracted from the Web pages browsed by the user. A page information extractor;
A product information extraction unit that extracts product information that is information about the product included in the page information from the page information extracted by the page information extraction unit;
From the page information extracted by the page information extraction unit, a summary generation unit that generates summary information that is a summary of information included in the page information;
A purchase behavior comprising: a statistical analysis unit that performs a predetermined statistical analysis process on the product information extracted by the product information extraction unit and the summary information generated by the summary generation unit Analysis equipment. The page information extraction unit specifies a Web page for purchasing a product as a Web page related to the product, specifies a Web page for browsing information about the product, and includes information included in the specified Web page. Extract some page information,
The product information extraction unit extracts the first product information that is the product information included in the Web page for purchasing the product, and is included in the Web page for browsing information about the product. Extracting the second product information which is product information;
The purchase behavior analysis apparatus according to claim 1, wherein the statistical analysis unit performs a predetermined analysis process on the summary information, the first product information, and the second product information.   The purchase behavior analyzing apparatus according to claim 1, wherein the acquisition unit acquires Web browsing data after personal information is deleted at the user terminal. Computer
Information included in a Web page related to a product among Web pages browsed by the user by executing an extraction process on Web browsing data including Web pages browsed by the user at a user terminal according to a predetermined extraction condition. A page information extraction process for extracting page information;
A product information extraction step of extracting product information that is information about the product included in the page information from the page information extracted by the page information extraction step;
From the page information extracted by the page information extraction step, a summary generation step for generating summary information that is a summary of information included in the page information;
Purchasing behavior comprising: a statistical analysis step for executing a predetermined statistical analysis process on the product information extracted by the product information extraction step and the summary information generated by the summary generation step Analysis method. On the computer,
Information included in a Web page related to a product among Web pages browsed by the user by executing an extraction process on Web browsing data including Web pages browsed by the user at a user terminal according to a predetermined extraction condition. A page information extraction procedure for extracting page information;
A product information extraction procedure for extracting product information, which is information about the product included in the page information, from the page information extracted by the page information extraction procedure;
A summary generation procedure for generating summary information that is a summary of information included in the page information from the page information extracted by the page information extraction procedure;
Purchasing behavior characterized by causing a statistical analysis procedure to execute a predetermined statistical analysis process on the product information extracted by the product information extraction procedure and the summary information generated by the summary generation procedure. Analysis program. A browsing unit for browsing web pages;
A user terminal having a transmission unit that transmits Web browsing data including a Web page browsed by the user in the browsing unit;
An acquisition unit for acquiring the web browsing data transmitted by the transmission unit of the user terminal;
By performing an extraction process on the Web browsing data acquired by the acquisition unit according to a predetermined extraction condition, page information that is information included in a Web page related to a product is extracted from the Web pages browsed by the user. A page information extractor;
A product information extraction unit that extracts product information that is information about the product included in the page information from the page information extracted by the page information extraction unit;
From the page information extracted by the page information extraction unit, a summary generation unit that generates summary information that is a summary of information included in the page information;
A server having a statistical analysis unit that executes a predetermined statistical analysis process on the product information extracted by the product information extraction unit and the summary information generated by the summary generation unit. Purchasing behavior analysis system. A computer as a user terminal
A browsing process for browsing web pages;
And transmitting a web browsing data including the web page browsed by the user in the browsing process,
A computer that acquires the Web browsing data transmitted by the transmitting step is
An acquisition step of acquiring the web browsing data transmitted by the transmission step of the user terminal;
By performing an extraction process on the Web browsing data acquired in the acquisition step according to a predetermined extraction condition, page information that is information included in a Web page related to a product is extracted from the Web pages browsed by the user. Page information extraction process;
A product information extraction step of extracting product information that is information about the product included in the page information from the page information extracted by the page information extraction step;
From the page information extracted by the page information extraction step, a summary generation step for generating summary information that is a summary of information included in the page information;
A control method comprising: a statistical analysis step of executing a predetermined statistical analysis process on the product information extracted by the product information extraction step and the summary information generated by the summary generation step. A browsing unit for browsing web pages;
A user terminal having a transmission unit that transmits Web browsing data including a Web page browsed by the user in the browsing unit;
An acquisition unit for acquiring the web browsing data transmitted by the transmission unit of the user terminal;
By performing an extraction process on the Web browsing data acquired by the acquisition unit according to a predetermined extraction condition, page information that is information included in a Web page related to a product is extracted from the Web pages browsed by the user. A page information extractor;
A product information extraction unit that extracts product information that is information about the product included in the page information from the page information extracted by the page information extraction unit;
From the page information extracted by the page information extraction unit, a summary generation unit that generates summary information that is a summary of information included in the page information;
A server having
A terminal having a statistical analysis unit that acquires the product information extracted by the product information extraction unit and the summary information generated by the summary generation unit from the server, and executes a predetermined statistical analysis process;
Purchasing behavior analysis system characterized by comprising A computer as a user terminal
A browsing process for browsing web pages;
And transmitting a web browsing data including the web page browsed by the user in the browsing process,
A server that acquires the Web browsing data transmitted in the transmission step,
An acquisition step of acquiring the web browsing data transmitted by the transmission step of the user terminal;
By performing an extraction process on the Web browsing data acquired in the acquisition step according to a predetermined extraction condition, page information that is information included in a Web page related to a product is extracted from the Web pages browsed by the user. Page information extraction process;
A product information extraction step of extracting product information that is information about the product included in the page information from the page information extracted by the page information extraction step;
Generating summary information that is a summary of information included in the page information from the page information extracted by the page information extraction step, and
A computer that acquires from the server the product information extracted by the product information extraction step and the summary information generated by the summary generation step,
A control method comprising a statistical analysis step of executing a predetermined statistical analysis process on the product information extracted by the product information extraction step and the summary information generated by the summary generation step.

Images