JPH11306160A - Service unit extraction method and device from service application history and recording medium recording extraction program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To automatically sort the services of many and unspecified servers by deciding the subsequent information address as a base point of service units if the different computer names are given to the preceding and subsequent addresses showing the information which are included in an information service application history and continuously referred to. SOLUTION: The application history data 21 are produced (S10) to include an information address consisting of a computer name, a directory name and a file name and an item that can identify whether a link is formed by hyperlink based on the information which is used right before the production of the data 21. Then the information addresses which are continuously referred to in the data 21 and preceding and subsequent to a link that is formed by hyperlink are acquired (S20). It's decided whether the computer names included in the said preceding and subsequent information addresses are coincident with each other (S30). If no coincidence is confirmed between both addresses, the computer name and the directory name which are included in the subsequent information address are extracted as a service unit 61 of the information service (S40).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for obtaining classification and statistics of network service usage survey data.

[0002]

2. Description of the Related Art In recent years, with the spread of WWW services and network services centered on the Internet, such as e-mail, to ordinary households, personal computers with a network communication function have been spreading.
For example, in the WWW service, a plurality of WWW server machines on the Internet have data such as characters, voices, and images in the machines, and the data is accessed and browsed from a browser software of an information terminal using a protocol called HTTP. Is done. Since these network services have value as advertising and marketing media, there is an increasing need to objectively know the usage of each service.

[0003] However, in the case of a service providing form using a large number of distributed machines, such as a WWW service, unlike other media such as televisions and newspapers, the number of service media is large, and the service media changes drastically. However, it is difficult to grasp the status of the service medium. Conventionally, W
In general, a means for analyzing the data history of the access log is used to check the usage status of the WW server.
The data history is a list in which data information names browsed by the WWW service are arranged in order of access time.

[0004] In order to indicate the location of data information and the name of the data information, generally, following a protocol name such as http: //, a URL describing the host name of a WWW server and the name of a directory or file in the server are provided. The so-called format is used. In the following description, the data information indicated by the URL is called a page for convenience. Therefore, analyzing the data history means classifying each page in the data history by service, that is, determining which service unit (site) each service belongs to.

FIG. 9 is a diagram for explaining the outline of a WWW server. As shown in the figure, the WWW server machine has a directory having a tree structure starting from a root directory. Each data is classified and stored in a tree-type directory in the server for each information field. For example, a hyperlink is provided from a directory A to a file b in a directory B that is not immediately below.

[0006] The unit of each service can be determined by a tree structure. FIG. 10 is an explanatory diagram of a service unit. According to the example shown in the figure, one WWW server has two contents, Mr. A's homepage and Mr. B's homepage. Each content corresponds to one service. In this case, from the viewpoint of organizing information, the WWW server creates separate directories for the two contents, and sets the information, that is, the pages, in the directories. Thus, one W
The WW server has two service contents, and each service is separated in a directory immediately below the root directory.

[0007] When information relating to the directory structure in the WWW server and information indicating in which directory unit the service provided by the WWW server is divided are given in advance, the data history and the advance By collating with a database holding given information, a method of classifying data history services is realized.

[0008] However, when the above information is not provided in advance, it is necessary for the analyst to examine the contents of the data history and to determine the unit of service separation. However, it is not easy to automate such a dividing method.
In a situation where the above information is not given in advance, a conventional method for automating the separation of services is to separate a range that can be clearly distinguished by a pattern, for example, WW
Classification is performed for each W server. When such a separation method is applied to the example in FIG. Y, two contents “Ms. A's homepage” and “Ms. B's homepage” in the same server are classified as the same service. Thus, the conventional method classifies separate services residing in the same server as the same service.

[0009]

The above-mentioned conventional service classification method has the following problems. 1. WWW in advance
In the case of a method of classifying services provided in the server by examining the directory structure in the server, the classifiable targets are limited to services in the WWW server. This method has a drawback that service classification cannot be performed on services in an unspecified number of WWW servers. 2. In addition, in order to perform service classification for an unspecified number of WWW server services by the conventional service classification method, verification of browsing data of WWW services is manually performed, or the unit of classification is determined by WWW server. It is necessary to do it in such a coarse unit. The manual verification is costly in terms of time and cost, and has a disadvantage that the range of the target WWW server is practically limited. Also, coarse classification such as by WWW server is extremely rare when it is effective as a service classification unit.

[0010] Accordingly, the present invention has been made in view of the above-described problems of the prior art, and has been made in view of the above-described problems. It is intended to provide a recording medium on which a program is recorded.

[0011]

In order to achieve the above object, a method and apparatus for extracting a service unit from a service use history according to the present invention focuses on the temporal context of a data history. According to the present invention, the computer names of the information addresses before and after indicating the continuously referenced information included in the use history of the information service (for example, the use history of the WWW service) referred to while tracing the information linked by the hyperlink are described. In a different method, a service unit is extracted from a service usage history, wherein a later information address is determined as a base point of the service unit.

FIG. 1 is a diagram for explaining the principle of the present invention. As shown in the figure, the method of extracting a service unit from the service usage history according to the present invention employs an information address including a computer name, a directory name, and a file name that provide information used in an information service, and the information immediately before the information address. The usage history data 21 including an item that can identify whether or not a link has been established by a hyperlink is created from the used information (step 10), and is continuously referred to in the usage history data 21, and The information addresses before and after the link by the hyperlink are acquired (step 20), and it is determined whether or not the computer names included in the acquired information addresses before and after match (step 30). If it is determined,
The computer name and the directory name included in the subsequent information address are converted into the service unit 61 in the information service.
(Step 40).

Further, the apparatus 2 for extracting a service unit from the service use history according to the present invention refers to an information server 6 for providing information, and is connected to the information server 6 and traces information linked by a hyperlink. In the information service system including the browser 4, the information address of the information used in the information service system and an item that can identify whether the information is linked by a hyperlink from the information used immediately before are described. A storage means 20 for storing usage history data, a means 10 for monitoring the usage status of the information by the browser 4, creating the usage history data, and storing it in the storage means 20; Means 3 for acquiring information addresses before and after being referred to and linked by a hyperlink Means 40 for comparing the acquired information addresses before and after to obtain whether or not the computer names included in the acquired information addresses before and after match, and a computer included in the information addresses after the acquired information addresses Means 50 for extracting a name and a directory name as a service unit in the information service system, and storing the extracted service name and directory name in the service unit data storage means 60.

[0014] A recording medium for recording a program for extracting a service unit from a service use history according to the present invention includes:
When the computer name of the information address before and after successively included in the use history of the information service referred to while tracing the information linked by the hyperlink is different, the subsequent information address is determined as the base point of the service unit. It is a recording medium that records a program for extracting a service unit from a service usage history, which is characterized by comprising a process.

In the recording medium of the present invention, an information address including a computer name and a directory name providing information used in the information service and a link from the information used immediately before by the information are hyperlinked. A process of acquiring information addresses before and after consecutively referred to in the usage history data including items that can identify whether or not the information addresses are linked by a hyperlink; and A process for determining whether or not the computer names included in the information service match; and a process for extracting the computer name and the directory name included in the subsequent information address as a service unit in the information service when it is determined that they do not match. Service unit based on service usage history A recording medium for recording a program to be extracted.

Further, the recording medium of the present invention includes an information address of information used in the information service system, and an item capable of identifying whether or not the information is linked by a hyperlink from information used immediately before. The process of acquiring the information addresses before and after consecutively referenced in the usage history data including and linked by the hyperlink matches the computer name included in the acquired information addresses before and after In order to determine whether or not the information address before and after the obtained information address is compared, a process of extracting the computer name and directory name included in the subsequent information address as a service unit in the information service system, Extract service unit from service usage history characterized by Which is a recording medium for recording the program.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described below in which items of a service unit are automatically extracted by focusing on the context of data history for each user. FIG. 3 is a configuration diagram of the service unit extraction system according to the first embodiment of this invention. A plurality of WWW servers 6 ₁ , 6 ₂ ,. . . Via the network 9, a plurality of user WWW browsers 4 ₁ , 4 ₂ ,. . . And the user enjoys the information service. Each user WWW browser 4 ₁ ,
42 _,. . . Includes a service unit extraction device 2 ₁ ,
2 ₂ ,. . . Is connected, and a service unit is extracted from the service usage history. Also, the service unit extraction device 2 ₁ ,
2 ₂ ,. . . Is connected via a network to a center server 8 including a database 9 for storing the extracted service units. The service unit extraction device and the user browser may be realized, for example, as a part of a so-called information terminal.

The service unit extraction devices 2 ₁ , 2 ₂ ,. . .
Is a URL (Uniform Resource Locator) that is continuously referenced from the usage history of an information service such as a WWW service that refers while tracing information linked by a hyperlink.
When the computer name such as the host name included in the information address such as tor) is different, the service unit is extracted from the service use history by regarding the moved information address as the base point of the service unit such as the site.

The features of the WWW service in this example are as follows. (A) In the WWW service, each data is connected in a network pattern by a hyperlink. The user is a certain WW
Information can be browsed one after another by following hyperlinks from the page on W. (B) On a WWW server, data generally takes a tree-type (directory) structure, and contents related to the same service are stored in the same node in the tree-type structure.

Referring now to FIG. 4 showing the directory structure of WWW server A and server B shown in the configuration diagram of FIG. 3, a specific example of service unit extraction according to the first embodiment of the present invention will be described. Will be described. In the following description, UR
Only files with the extension html are handled as L files. FIG. 5 is a flowchart of a service unit extraction process according to the first embodiment of the present invention.

Step 100: The user is the WWW server A
Browse the file (e.html) that exists on the root directory (/) on (URL name http://www.a.com/). Step 102: At this time, when the user enters http://www.a.com/
The fact that the user is browsing the page e.html is described in the history data in URL notation. Step 104: The user follows the hyperlink on e.html and copies another file (f.html) existing in the directory (/ job /) on the same server A (http://www.a.com/). browse.

Step 106: At this time, the user enters ht
You have visited the page tp: //www.a.com/job/f.html and it is recorded on the history data. At this time, the history data is recorded in chronological order, such as 1. http://www.a.com/e.html 2. http://www.a.com/job/f.html hyper . Where hype
The letter r means that the transition from History 1 to History 2 is e.html
It is provided for convenience to indicate the fact that it was done by the hyperlink above.

Step 108: Here, since the previous history 1 and the subsequent history 2 are connected by a hyperlink, the server names included in the URLs of the preceding and succeeding histories are compared. Step 110: At this time, both server names are http: /
Since it is /www.a.com/ and they match, it is determined that there is no change in the browsing server. Since the directory is moved within the same server, the directory of the history 2 is not extracted as a service unit.

Step 112: Next, the user enters f.html
Follow the hyperlink above to another server B (URL name
http://www.b.co.jp/) directory (/ musi
Browse the file g.html on c /). Step 114: At this time, http://www.b.co.jp/music
Since you have viewed the page /g.html, the history data is 1. http://www.a.com/e.html 2. http://www.a.com/job/f.html hyper 3. Updated as http://www.b.co.jp/music/g.html hyper.

Step 116: History 2 before and History 3 after
Is linked by a hyperlink, so U
The server names included in the RL are compared. Step 118: At this time, the server name of the previous history 2 is
http://www.a.com/ and the server name of History 3 is
Since the two do not match at http://www.b.co.jp/, it is determined that the browsing server has changed.

Step 120: Since the directory is moved between different servers, the information address of the history 3 http:
//www.b.co.jp/music/g.html to file name g.html
URL except for http://www.b.co.jp/music/
The lower page is extracted as a service unit and transmitted to the center server. Step 122: The center server is connected to http: //www.b.co.j
Register p / music / in the service unit database.

A similar process is repeated for histories 3 to 4, histories 4 to 5 and histories 5 to 6. Step 124: Next, the user follows the hyperlink on the file g.html and the same server B (http: //www.bc
Browse another file (h.html) in the directory (/ suzuki / books /) on o.jp/).

Step 126: At this time, the user enters ht
This means that you have visited the page tp: //www.b.co.jp/suzuki/books/h.html and it is recorded on the history data. At this time, the history data is: 1.http: //www.a.com/e.html 2.http: //www.a.com/job/f.html hyper 3.http: //www.b. co.jp/music/g.html hyper 4. Updated as http://www.b.co.jp/suzuki/books/h.html hyper.

Step 128: Here, since the preceding history 3 and the following history 4 are connected by a hyperlink, the server names included in the URLs of the preceding and succeeding histories are compared. Step 130: At this time, both server names are http: /
/www.b.co.jp/, which matches, it is determined that there is no change in the browsing server. Since the directory is moved within the same server, the directory of the history 4 is not extracted as a service unit.

Step 132: Next, the user enters the file
Following the hyperlink on h.html, the same server B (ht
Browse another file (i.html) in the directory (/ suzuki /) on tp: //www.b.co.jp/). Step 134: At this time, the user goes to http: //www.bc
The page "o.jp/suzuki/i.html" is browsed and recorded on the history data. At this time, the history data is: 1.http: //www.a.com/e.html 2.http: //www.a.com/job/f.html hyper 3.http: //www.b. co.jp/music/g.html hyper 4.http: //www.b.co.jp/suzuki/books/h.html hyper 5.http: //www.b.co.jp/suzuki/i. Updated like html hyper.

Step 136: Here, since the preceding history 4 and the following history 5 are connected by a hyperlink, the server names included in the URLs of the preceding and succeeding histories are compared. Step 138: At this time, both server names are http: /
/www.b.co.jp/, which matches, it is determined that there is no change in the browsing server. Since the directory is moved within the same server, the directory of the history 5 is not extracted as a service unit.

Step 140: Next, the user enters i.html
Follow the above hyperlink to another server A (URL name
http://www.a.com/) directory (/ sports
Browse the file j.html on / baseball). Step 142: At this time, http://www.a.com/sports/
Since you have visited the page called baseball / j.html, the history data is 1. http://www.a.com/e.html 2. http://www.a.com/job/f. html hyper 3.http: //www.b.co.jp/music/g.html hyper 4.http: //www.b.co.jp/suzuki/books/h.html hyper 5.http: // www.b.co.jp/suzuki/i.html hyper 6. Updated as http://www.a.com/j.html hyper.

Step 144: History 5 before and History 6 after
Is linked by a hyperlink, so U
The server names included in the RL are compared. Step 146: At this time, the server name of the previous history 5 is
http://www.b.co.jp/
Since the two do not match at http://www.a.com/, it is determined that the browsing server has changed.

Step 148: Since the directory is moved between different servers, the information address of the history 6 http:
// File name from www.a.com/sports/baseball/j.html
URL except for j.html http://www.a.com/sp
Extract pages lower than orts / baseball / as service units and register them in the service unit database of the center server.

With the above processing, the service unit registered in the database of the center server is http://www.b.co.jp/music/ http://www.a.com/sports/baseball/. .

The comparison of the server name for determining whether or not to extract as a service unit can be realized by a character string comparison between the server name of the preceding URL and the server name of the subsequent URL. The operation of removing the file name for extracting the service unit from the URL is realized by removing a part of the character string. To summarize an example of the service unit extraction processing of the present example, the following two conditions are satisfied. A hyperlink has been taken to the current browse page from the previous browse page. 2. The WWW server is different between the last browse page and the current browse page. Is satisfied, the directory of the current browse page is extracted as a service unit.

In this embodiment, if the current browse page is stored in another node in the same server as the immediately preceding browse page, the current browse page is not extracted as a service unit. This is because a general WWW service takes into account the fact that "a case where contents of the same service are stored in different nodes exists to an extent that cannot be ignored". If the current browse page is stored in another node on the same server as the previous browse page, extracting the current browse page as a service unit not only extracts content units that need to be separated, but also content that does not need to be separated. Units are also extracted at the same time, and contents that do not need to be divided tend to be more than contents that need to be divided. However, if necessary, when the current browse page is stored in another node in the same server as the previous browse page, it is easy to change the current browse page to be extracted as a service unit.

When the immediately preceding browse page and the current browse page are in the same server and the same node, there is a high possibility that the content is the same, so the current browse page is not extracted as a service unit. In the configuration of the first embodiment of the present invention, a service unit extraction device is provided for each user browser. However, it is also possible to configure so that only one service unit extraction device is provided in the service unit extraction system. FIG. 6 is a configuration diagram of a service unit extraction system according to the second embodiment of the present invention in which the service unit extraction device 2 is provided in the center server 8. The service unit extraction processing according to the second embodiment of the present invention operates in the same manner as the service unit extraction processing described with respect to the first embodiment of the present invention.

In the description of the above embodiment, the service unit extraction process is such that when a user refers to information while tracing information linked by a hyperlink, the service unit is simultaneously extracted from the history and the extracted service unit is extracted. The unit is notified to the center server. However, the present invention is not limited to such an example. For example, service units extracted in real time may be collectively notified to the center server later, or the user history data created in advance may be used. On the other hand, a service unit may be extracted.

[0040]

FIG. 7 is a schematic diagram of a user history collection system to which a method and apparatus for extracting a service unit from a use history according to the present invention are applied. In this system, the information terminal 100 is an actual user's machine, and the information servers 121, 122, 123. . . A WWW browser 102 is provided for use of services from the Internet. By providing the information terminal 100 with a monitoring / history information creation unit 104 for monitoring the user's WWW browser 102,
Monitor the behavior of the WWW browser. By monitoring the WWW browser, it is possible to determine whether the obtained user history is obtained by a hyperlink or a bookmark move.

The user history acquired by the monitoring / history information creation unit 104 is sent to the user history management unit 112 of the center server 110 via the network 130 and stored in the user history database 114. The user history data stored in the user history database 114 is used, for example, for analysis of user behavior, audience rating survey on the WWW, and the like through an analysis process 210 such as statistical processing.

The method and apparatus for extracting the service unit from the usage history according to the present invention are applied to a part 200 of the preprocessing stage of the analysis process of the user history data. FIG. 8 is a schematic explanatory diagram of a service classification system applied to the preprocessing stage of the analysis process. The service unit (candidate) is extracted by applying the service unit extraction process from the usage history according to the present invention to the user history (or user log history) extracted from the database 114 (step 200). Next, of the service units obtained in step 200, unnecessary service units are deleted (step 202), and rules relating to the user history are created for each URL (step 204). The rules thus created are applied to the user history, and an item called a service unit is added to the user history (step 206). Finally, the user history to which the item of the service unit is added is written back to the database 114, so that the user history database is updated and used for a further analysis process.

The system described with reference to FIGS. 7 and 8 is configured to process a previously collected user history, but incorporates a process of extracting a service unit together with a monitoring process in a user's information terminal. Alternatively, information of service units extracted and processed in real time on the information terminal side may be collected later in the center server. Further, the service unit obtained in real time may be combined with the classification process and configured to provide feedback to the user. Thus, for example, it is possible to notify the user on the screen whether or not the service that the user is currently browsing is a new service unit that has not been browsed so far.

Further, the configuration of the service unit extraction system from the usage history is not limited to the example described in the above embodiment, and each component requirement of the service unit extraction system is constructed by software (program). It is also possible to extract the service unit by storing it in a disk unit or the like and installing it on a computer of the service unit extraction system as needed. Further, the constructed program can be stored in a portable recording medium such as a floppy disk or a CD-ROM, and can be used for general purposes in a case where such a system is used.

The present invention is not limited to the above embodiment, but can be variously modified and applied within the scope of the claims.

[0046]

As described above, according to the present invention,
Services are provided by a large number of distributed machines such as WWW services, the number of service media is large, and even when the service media changes drastically, service units are automatically extracted from the user's service usage history. So
Based on the extracted service units, the status of service media having value as an advertisement / marketing media can be grasped, and the usage status of each service can be objectively known.

[Brief description of the drawings]

FIG. 1 is a diagram illustrating the principle of the present invention.

FIG. 2 is a configuration diagram of a service unit extraction device of the present invention.

FIG. 3 is a configuration diagram of a service unit extraction system according to the first embodiment of this invention.

FIG. 4 is an explanatory diagram of a specific example of service unit extraction according to the first embodiment of the present invention.

FIG. 5 is a processing flowchart of service unit extraction according to the first embodiment of the present invention.

FIG. 6 is a configuration diagram of a service unit extraction system according to a second embodiment of this invention.

FIG. 7 is a schematic diagram of a user history collection system.

FIG. 8 is a schematic explanatory diagram of a service unit classification system.

FIG. 9 is a schematic explanatory diagram of a WWW service.

FIG. 10 is an explanatory diagram of a service unit.

[Explanation of symbols]

 4 Browser 6 Information Server 10 Usage History Data Creation Means 20 Usage History Data Storage Means 30 Information Address Acquisition Means 40 Computer Name Acquisition Means 50 Service Unit Extraction Means 60 Service Unit Data Storage Means

────────────────────────────────────────────────── ─── Continued on the front page (51) Int.Cl. ⁶ Identification code FIG06F 15/419 320 (72) Inventor Ken Motobashi 3-19-2 Nishishinjuku, Shinjuku-ku, Tokyo Nippon Telegraph and Telephone Corporation

Claims (7)

[Claims] When the computer name of an information address before and after consecutively referenced information included in a use history of an information service referred to while tracing information linked by a hyperlink is different, the subsequent information address is assigned to a service unit. A method of extracting a service unit from a service usage history, characterized in that the service unit is determined as a base point of the service. 2. An information address including a computer name and a directory name for providing information used in the information service, and an item for identifying whether or not the information was linked by a hyperlink from information used immediately before. The information addresses before and after that are successively referenced in the usage history data including and linked by a hyperlink are acquired, and the computer names included in the acquired information addresses before and after match. It is determined whether or not they match, and if they do not match, the computer name and directory name included in the subsequent information address are extracted as service units in the information service, and the service unit is extracted from the service use history. how to. 3. Usage history data including an information address of information used in the information service system and an item that can identify whether the information has been linked by a hyperlink from information used immediately before is stored. Means for acquiring information addresses before and after successively referred to in the usage history data and linked by a hyperlink, and included in the acquired information addresses before and after Means for comparing the acquired information addresses before and after to determine whether or not the computer names match, and using the computer name and the directory name included in the information address after the above as a service unit in the information service system Means for extracting and storing the service from the service usage history. A unit that extracts unit. 4. The apparatus for extracting a service unit from a service use history according to claim 3, further comprising means for monitoring a state in which information is used, creating the use history data, and storing the data in the storage means. 5. When the computer name of an information address before and after a continuous reference included in the use history of an information service referred to while tracing information linked by a hyperlink is different, a subsequent information address is used as a service unit. A recording medium for recording a program for extracting a service unit from a service use history, characterized by comprising a process of determining the service unit as a base point. 6. An information address including a computer name and a directory name for providing information used in the information service, and an item for identifying whether the information was linked by a hyperlink from information used immediately before. The process of acquiring information addresses before and after consecutively referenced in the usage history data including and linked by a hyperlink, and the computer name included in the acquired information addresses before and after match And a process of extracting a computer name and a directory name included in the subsequent information address as a service unit in the information service when it is determined that they do not match. Write a program to extract service units from service usage history The recording medium. 7. Usage history data including an information address of information used in an information service system and an item for identifying whether or not the information has been linked by a hyperlink from information used immediately before. A process of acquiring information addresses before and after successively referred to and linked by a hyperlink, and to determine whether or not the computer names included in the acquired information addresses before and after match. Using the service, characterized by comprising: a process of comparing the obtained information addresses before and after the obtained information address; and a process of extracting a computer name and a directory name included in the obtained information address as a service unit in the information service system. A record of a program that extracts service units from the history Media.