JP2005070889A - Attribute judgment system, method, and computer program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an attribute judgment system for judging the attributes of a user according to access history information to a Web site without making it necessary to receive personal information from a user. <P>SOLUTION: The attribute judgment system for judging attributes showing the characteristics of a user is provided with a storage means for storing the sample data of each of media showing the user configuration ratio of each attribute and the access history information of the user to the media, a history managing means for writing the access history information in the storage means when the terminal equipment of the user has performed access to the media and an attribute judging means for judging the attributes of the user from the access history information and sample data in the storage means. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

【００２８】
上記の式により、ユーザ１０が男性である確率は約７６％と算出される。属性判定手段３１４は、このユーザ１０の属性は男性である確率が高いと判定し、ユーザ属性情報に男性の属性を書き込むとともに、この確率に応じて信頼性の情報に信頼性Ｂを書き込む。属性判定手段３１４は、同様の方法によりこのユーザ１０の他の属性を判定し、信頼性の情報とともにデータベース３２内のユーザ属性情報に書き込む。
【００２９】
分析サーバ３１の属性判定手段３１４は、上記のようにメディアのサイト２０へアクセスがある度に、ユーザの履歴情報に記録されている全てのメディアに対応したサンプルデータをデータベース３２から読出し、各属性に対してこのユーザが当該属性を有する確率を求める。そして、この確率から想定されるユーザの属性の情報と、その信頼性の情報によりデータベース３２内のユーザ属性情報を更新する手順を繰り返す。
【００３０】
なお、出力手段３１５は、ユーザがある属性を有する確率が属性を判断できない平準化された所定の範囲、例えば、男性の属性である確率４５〜５５％であることを配信条件として、性別などのユーザ情報（属性）を問い合わせるアンケートの画像情報（画面情報）を直接端末装置１１へあるいはメディア配信サーバ２１を介して端末装置１１へ配信してもよい。分析サーバ３１は、端末装置１１がブラウザに表示するアンケートの画面にユーザ１０が入力したユーザ情報（属性）を取得し、データベース３２内のユーザ属性情報に書き込む。
また、ユーザがある属性を有する確率を求めるときに、各メディアを利用した回数を加味してもよい。
【００３１】
なお、上記の実施の形態において、ユーザＩＤを端末装置１１内のＣｏｏｋｉｅから取得しているが、ユーザ１０が端末装置１１から各メディアのサイト２０にアクセスしたときに、端末装置１１へユーザＩＤの入力を促すユーザＩＤ入力画面をブラウザに表示させることにより、ユーザ１０が入力したユーザＩＤを取得することでもよい。
また、分析サーバ３１からサイト２０へユーザ属性情報を通知してもよい。
また、端末装置１１は、双方向通信が可能なテレビなどであってもよい。このとき配信されるメディアは、テレビ放送などである。
【００３２】
なお、上述の端末装置１１、メディア配信サーバ２１、データベース２２、分析サーバ３１、及び、データベース３２は、内部にコンピュータシステムを有している。そして、上述した動作の過程は、プログラムの形式でコンピュータ読み取り可能な記録媒体に記憶されており、このプログラムをコンピュータシステムが読み出して実行することによって、上記処理が行われる。ここでいうコンピュータシステムとは、ＯＳや周辺機器等のハードウェアを含むものである。
【００３３】
また、「コンピュータ読み取り可能な記録媒体」とは、ＲＯＭの他に、磁気ディスク、光磁気ディスク、ＣＤ−ＲＯＭ、ＤＶＤ−ＲＯＭ等の可搬媒体、コンピュータシステムに内蔵されるハードディスク等の記憶装置のことをいう。さらに「コンピュータ読み取り可能な記録媒体」とは、インターネット等のネットワークや電話回線等の通信回線を介してプログラムが送信された場合のシステムやクライアントとなるコンピュータシステム内部の揮発性メモリ（ＲＡＭ）のように、一定時間プログラムを保持しているものも含むものとする。
【００３４】
また、上記プログラムは、このプログラムを記憶装置等に格納したコンピュータシステムから、伝送媒体を介して、あるいは、伝送媒体中の伝送波により他のコンピュータシステムに伝送されてもよい。ここで、プログラムを伝送する「伝送媒体」は、インターネット等のネットワーク（通信網）や電話回線等の通信回線（通信線）のように情報を伝送する機能を有する媒体のことをいう。
また、上記プログラムは、前述した機能の一部を実現するためのものであっても良い。さらに、前述した機能をコンピュータシステムにすでに記録されているプログラムとの組み合わせで実現できるもの、いわゆる差分ファイル（差分プログラム）であっても良い。
【００３５】
【発明の効果】
この発明によれば、ユーザのメディアへのアクセス履歴情報と、各メディアから提供される属性毎のユーザ構成比率の情報を基に、ユーザから個人的な情報の提供を受けることなく、ユーザ属性を判定することができる。これにより、各メディアは、ユーザ情報を登録していないユーザに対しても、想定した属性に従った広告などを配信することができるため、広告の効果を高めることができる。
【図面の簡単な説明】
【図１】この発明の一実施の形態による属性判定システムを用いたウェブシステムの構成を示すブロック図である。
【図２】同実施の形態による属性判定システムの内部構成を示すブロック図である。
【図３】同実施の形態による属性判定システムの動作フローを示す図である。
【図４】同実施の形態によるユーザの属性の判定の方法を説明するための図である。
【符号の説明】
１０…ユーザ
１１…端末装置
２０ａ、２０ｂ、２０ｎ…サイト
２１ａ、２１ｂ、２１ｎ…メディア配信サーバ
２２ａ、２２ｂ、２２ｎ、３２…データベース
３０…属性判定システム
３１…分析サーバ
３１１…制御手段
３１２…通信手段
３１３…履歴管理手段
３１４…属性判定手段
３１５…出力手段[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an attribute determination system and method for determining an attribute of a user who accesses a website, and a computer program.
[0002]
[Prior art]
In recent years, with the spread of the Internet, advertisers and advertisement distributors have devised to increase the advertising effect by distributing highly-preferred advertisements suitable for attributes such as sex and preference of each user. For example, Patent Document 1 describes a technology for adding appropriate advertisement information in content distribution by receiving a content distribution request from a user terminal and reading out advertisement data to be added according to the user information of the user from a database. Has been.
[0003]
[Patent Document 1]
Japanese Patent Laid-Open No. 2002-203119
[Problems to be solved by the invention]
Conventionally, user information (user attributes) used to determine advertisement information to be distributed has been collected and registered in a database by a company that provides media by conducting a questionnaire on paper or online. This was a cumbersome task for the user or the company. Moreover, personal information is included in the user information, and the user often hesitates to register, and the user information cannot be shared by a plurality of companies. Therefore, the method in which each company distributes advertisements corresponding to attributes only to users who manage the user information itself is insufficient for expanding the users more widely.
[0005]
The present invention has been made in consideration of such circumstances, and its purpose is to determine the attribute of this user based on the access history to the website without receiving personal information from the user. An object attribute determination system and method, and a computer program are provided.
[0006]
[Means for Solving the Problems]
The present invention has been made to solve the above-mentioned problems, and the invention according to claim 1 is an attribute determination system for determining an attribute indicating a feature of a user, each of which indicates a user composition ratio for each attribute. Storage means for storing sample data of media and access history information of the user to the medium; history management means for writing access history information to the storage means when a user terminal device accesses the medium; An attribute determination system comprising: attribute determination means for determining an attribute of the user from access history information and sample data in the storage means.
[0007]
The invention according to claim 2 is the attribute determination system according to claim 1, wherein the attribute determination means determines the attribute by calculating a probability that the user has the attribute in the sample data. It is characterized by.
[0008]
Invention of Claim 3 is the attribute determination system of Claim 2, Comprising: Furthermore, the information which collects user information from the said terminal device when the probability calculated by the said attribute determination means exists in the predetermined range It is characterized by comprising a collecting means.
[0009]
The invention according to claim 4 is the attribute determination system according to any one of claims 1 to 3, and further outputs a medium corresponding to the attribute of the user determined by the attribute determination means. An output means is provided.
[0010]
The invention according to claim 5 is an attribute determination method for determining an attribute indicating a feature of a user, wherein sample data of each medium indicating a user composition ratio for each attribute, access history information of the user to the medium, Storage means is stored, and when the user's terminal device accesses the medium, the access history information is written to the storage means. From the access history information and sample data in the storage means, the attribute of the user is written. It is an attribute determination method characterized by determining.
[0011]
The invention according to claim 6 is a computer program used in an attribute determination system that determines an attribute indicating a user's characteristics, and includes sample data of each medium indicating a user composition ratio for each attribute, and the user's medium. Writing the access history information to the storage means for storing the access history information when the user's terminal device accesses the medium, and the user's attribute from the access history information and sample data in the storage means And a step of determining a computer program.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
Embodiments of the present invention will be described below with reference to the drawings.
FIG. 1 is a block diagram showing a configuration of a web system using an attribute determination system according to an embodiment of the present invention.
[0013]
The terminal device 11 is, for example, a personal computer or a mobile terminal that can be connected to the Internet owned by the user 10, and is connected to a network 40 including a public telephone network, a mobile phone network, an Internet network, a private network, and the like. Then, screen information provided from the sites 20a to 20n (hereinafter simply referred to as “site 20”) of the media A to N (hereinafter simply referred to as “media”) is displayed on a display, or by a mouse or a keyboard. A browser or the like for transmitting data input to the screen information to the site 20 is provided.
[0014]
Media sites 20a to 20n such as ISPs (Internet Service Providers) and general companies that are affiliates are respectively media distribution servers 21a to 21n (hereinafter referred to as "media distribution server 21") and databases 22a to 22n (hereinafter referred to as " Database 22 ”). The site 20 is connected to the terminal device 11 and the attribute determination system 30 via the network 40.
The media distribution server 21 provides screen information for displaying various contents (media) of a website such as a home page to the terminal device 11 accessed via the network 40. This screen information includes image information provided from the attribute determination system 30. For example, screen information for displaying content is described in HTML (hypertext markup language), XML (extensible markup language), or the like. The image information provided from the attribute determination system 30 is a flash image, a GIF animation image, or the like for displaying a banner advertisement or the like. Further, the media distribution server 21 may distribute video or the like as content (media).
The database 22 stores content data to be provided to the terminal device 11 and sample data indicating the composition ratio of the user 10 for each attribute in the site 20. The attribute is information indicating the characteristics of the user, for example, information indicating gender, region, occupation, age, consumer preference / hobby, behavior tendency on the Internet or a specific site, and the like. Details of the sample data will be described later.
[0015]
FIG. 2 is a block diagram in which the inside of the attribute determination system 30 held by an advertising agency as an affiliate service provider is developed, and only functional blocks related to the present invention are extracted and shown. The attribute determination system 30 includes an analysis server 31 and a database 32.
[0016]
The database 32 stores sample data, access history information of the user 10, user attribute information indicating the determined attribute of the user 10, and media distribution information.
The sample data includes media identification information, the number of samples, and information on the user composition ratio for each attribute. An example of sample data is shown below.
Media identification information: Media A URL (Universal Resource Locator)
-Number of samples: 1000-User composition ratio for each attribute:
-Gender: 80% male, 20% female
-The area where I live: Hokkaido 5%, Tohoku 5%, ...
-Consumer preference: (Car) very interested 30%, rather 25% interested, ... (music) 25% interested in Western music, 10% interested in classical music, ... ( …)…
-Internet behavior trends: (Online shopping) 10% frequently used, 15% frequently used, ..., (...) ...
-Trends in behavior at online shopping sites: (Discount product sites) 5% to browse, 10% to browse often, 20% to browse sometimes, ... (Purchase of limited quantity products) 5% to purchase often, sometimes purchase 15% ... (...) ...
-Age: 15% in teens, 20% in 20s, ...
[0017]
The access history information includes identification information (user ID) of the user 10, identification information (URL and the like) of the site 20 accessed from the terminal device 11, and information on the date and time of access.
The user attribute information includes identification information (user ID) of the user 10 and information of the determined attribute (for example, gender: male, area of residence: Kinki, consumer preference: very interested in cars, on the Internet Behavior tendency: frequently used online shopping, etc.), and reliability information of each attribute (for example, ranks of A, B,...).
The media distribution information includes image information (media) for displaying various contents such as advertisements, notifications, questionnaires, and present applications, and distribution conditions indicating attributes of the user 10 to which the image information should be distributed. Note that the distribution condition may be a combination of a plurality of attributes, or may be a condition in which other information such as time and access source media is added as a condition. Alternatively, one distribution condition may be associated with a plurality of pieces of image information, and image information may be selected according to a predetermined ratio or priority.
[0018]
The analysis server 31 includes a control unit 311, a communication unit 312, a history management unit 313, an attribute determination unit 314, and an output unit 315.
The control unit 311 includes a CPU (central processing unit) and various memories, and controls each unit, temporarily stores data, transfers data, and the like.
The communication unit 312 is connected to the network 40 through a dedicated line, a broadband line, or the like, and transmits / receives data to / from the terminal device 11, the media distribution server 21, and the database 22. In addition, data is transmitted to and received from the database 32 via a LAN (Local Area Network) or the network 40.
The history management unit 313 has a function of accumulating user 10 access history information in the database 32.
The attribute determination unit 314 has a function of determining the attribute of the user 10 based on the sample data of each medium in the database 32 and the access history information of the user 10, and registering / updating the user attribute information in the database 32.
The output unit 315 has a function of reading image information corresponding to the attribute of the user 10 determined by the attribute determination unit 314 from the media distribution information in the database 32 and returning it to the media distribution server 21.
[0019]
Next, the operation of the web system will be described with reference to the operation flow of the attribute determination system 30 in FIG.
Each media site 20 always registers the latest sample data in its own media in each database 22. This sample data is a result collected by the site 20 of each media displaying a questionnaire screen including a question and a button for selecting an answer on the browser of the terminal device 11 and obtaining an answer input by the user 10. Is calculated and registered. Alternatively, sample data obtained from a questionnaire result collected by each media using a questionnaire sheet or the like may be directly registered in the database 22 by an input means (not shown) such as a keyboard. The analysis server 31 of the attribute determination system 30 acquires sample data from the database 22 of each medium periodically or whenever sample data is updated, and accumulates the sample data in the database 32.
[0020]
Here, it is assumed that the user 10 accesses the site 20 a of the media A from the terminal device 11. Specifically, when the user 10 designates the URL of the site 20a with a browser displayed on the terminal device 11, content is requested to the media distribution server 21a. When the media distribution server 21a receives a request for content, in order to generate screen information for displaying the requested content, the content data is read from the database 22a, and image information embedded in the screen information (or like a pop-up) The screen information to be displayed at the same time is requested to the analysis server 31. Here, it is assumed that this is the first access from the user 10 to the sites 20 of the media A to N.
[0021]
The analysis server 31 receives a request for image information from the media distribution server 21a. Thereby, the history management unit 313 recognizes that the terminal device 11 (user 10) has accessed the site 20a of the media A (step S1). Then, the history management unit 313 reads a cookie in the terminal device 11 and determines whether or not it is the first access from the terminal device 11 to the sites 20 of the media A to N (step S2). If the cookie does not exist, it is determined that the access is the first access, a user ID is generated, and the cookie in which this user ID is set is stored in the terminal device 11 (step S3).
Cookie is a technology that provides a mechanism for a site provider to temporarily write and store data on a user terminal device through a browser. Information about the user, the date and time when the site was last visited, the site You can record the number of visits.
[0022]
The history management unit 313 of the analysis server 31 writes the identification information of the medium A and the access date / time information in the access history information in the database 32 in association with the user ID (step S4). Subsequently, the attribute determination unit 314 reads the sample data corresponding to the medium A indicated in the access history information of the user 10 from the database 32, determines the attribute of the user 10, and the result is the user attribute information in the database 32. (Step S5).
[0023]
Next, the output unit 315 reads the image information corresponding to the distribution condition matching the attribute determined by the attribute determination unit 314 from the media distribution information in the database 32, and returns it to the media distribution server 21a of the medium A (step S6). ). The media distribution server 21 a generates screen information for displaying the image information received from the analysis server 31 and the content data read from the database 22 a, and returns the screen information to the terminal device 11. The browser of the terminal device 11 displays images such as contents and advertisements in accordance with the screen information received from the media A site 20a.
[0024]
Thereafter, the same procedure as described above is performed when the terminal site 11 of the user 10 accesses the media site 20. However, in step S <b> 2, the history management unit 313 of the analysis server 31 acquires the user ID from the cookie read from the terminal device 11. As a result, it is determined that this is not the first access from the terminal device 11 to the media site 20, and step S4 for updating the access history information in the database 32 is executed. Subsequently, in step S5, the attribute determination unit 314 of the analysis server 31 reads the sample data corresponding to all the media indicated in the updated access history information of the user 10 from the database 32 and determines the attribute of the user 10. The result is written in the user attribute information in the database 32.
[0025]
FIG. 4 is a diagram for explaining a method of determining a user attribute in step S5 of FIG.
For example, in the sample data of media A registered in the database 32 of the attribute determination system 30, the number of samples is 1000, and the composition ratio of the user with respect to gender (attribute) is set to 80% male and 20% female. It is assumed that the sample data of the media B is set with 800 samples and 70% male and 30% female as the user composition ratio with respect to gender.
[0026]
Here, when the terminal device 11 accesses the site 20a of the media A, and this is the first access to the site 20 of the media A to N, the history information includes only access to the site 20a of the media A. Is recorded. For this reason, the attribute determination unit 314 reads the user composition ratio for each attribute from only the sample data of the medium A. The attribute determination means 314 determines that the attribute of the user 10 is male because the composition ratio of the user with respect to the read gender (attribute) indicates “male 80%, female 20%”. The male is written in the user attribute information in 22. Further, according to this probability, for example, the reliability information of this attribute is written as reliability A if the probability is 80% or more, reliability B if the probability is less than 80% and 70% or more, and so on.
[0027]
Next, it is assumed that the terminal device 11 accesses the site 20b of the media B. At this time, the attribute determination unit 314 of the analysis server 31 reads the sample data of the media A and the media B according to the access history information of the user 10. Then, the probability that the user 10 is male is determined by the following equation.

[0028]
From the above formula, the probability that the user 10 is male is calculated to be about 76%. The attribute determination unit 314 determines that the attribute of the user 10 has a high probability of being male, writes the male attribute in the user attribute information, and writes the reliability B in the reliability information according to the probability. The attribute determination unit 314 determines other attributes of the user 10 by a similar method, and writes the attributes in the user attribute information in the database 32 together with the reliability information.
[0029]
The attribute determination means 314 of the analysis server 31 reads sample data corresponding to all media recorded in the user's history information from the database 32 every time the media site 20 is accessed as described above. The probability that this user has the attribute is obtained. And the procedure which updates the user attribute information in the database 32 with the information of the user attribute assumed from this probability and the reliability information is repeated.
[0030]
Note that the output means 315 uses a distribution condition that the probability that the user has an attribute is a predetermined level in which the attribute cannot be determined, for example, the probability of being a male attribute is 45% to 55%. Image information (screen information) of a questionnaire for inquiring user information (attributes) may be distributed directly to the terminal device 11 or to the terminal device 11 via the media distribution server 21. The analysis server 31 acquires the user information (attribute) input by the user 10 on the questionnaire screen displayed on the browser by the terminal device 11 and writes it in the user attribute information in the database 32.
Further, when the probability of having a certain attribute is obtained, the number of times each medium is used may be taken into account.
[0031]
In the above embodiment, the user ID is acquired from the cookie in the terminal device 11. However, when the user 10 accesses the site 20 of each medium from the terminal device 11, the user ID is transmitted to the terminal device 11. The user ID input by the user 10 may be acquired by causing the browser to display a user ID input screen for prompting input.
Further, the user attribute information may be notified from the analysis server 31 to the site 20.
Further, the terminal device 11 may be a television or the like capable of bidirectional communication. The media distributed at this time is a television broadcast or the like.
[0032]
Note that the terminal device 11, the media distribution server 21, the database 22, the analysis server 31, and the database 32 described above have a computer system therein. The above-described operation process is stored in a computer-readable recording medium in the form of a program, and the computer system reads and executes this program to perform the above processing. The computer system here includes an OS and hardware such as peripheral devices.
[0033]
In addition to ROM, “computer-readable recording medium” refers to a portable medium such as a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, or a storage device such as a hard disk built in a computer system. That means. Furthermore, the “computer-readable recording medium” refers to a volatile memory (RAM) inside a computer system that becomes a client or a system when a program is transmitted via a network such as the Internet or a communication line such as a telephone line. In addition, those holding programs for a certain period of time are also included.
[0034]
The program may be transmitted from a computer system storing the program in a storage device or the like to another computer system via a transmission medium or by a transmission wave in the transmission medium. Here, the “transmission medium” for transmitting the program refers to a medium having a function of transmitting information, such as a network (communication network) such as the Internet or a communication line (communication line) such as a telephone line.
The program may be for realizing a part of the functions described above. Furthermore, what can implement | achieve the function mentioned above in combination with the program already recorded on the computer system, and what is called a difference file (difference program) may be sufficient.
[0035]
【The invention's effect】
According to the present invention, the user attributes can be obtained without receiving personal information from the user based on the access history information of the user to the medium and the information of the user composition ratio for each attribute provided from each medium. Can be determined. Thereby, since each medium can distribute the advertisement etc. according to the attribute assumed also to the user who has not registered user information, the effect of an advertisement can be heightened.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of a web system using an attribute determination system according to an embodiment of the present invention.
FIG. 2 is a block diagram showing an internal configuration of the attribute determination system according to the embodiment;
FIG. 3 is a diagram showing an operation flow of the attribute determination system according to the embodiment;
FIG. 4 is a diagram for explaining a user attribute determination method according to the embodiment;
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 10 ... User 11 ... Terminal device 20a, 20b, 20n ... Site 21a, 21b, 21n ... Media distribution server 22a, 22b, 22n, 32 ... Database 30 ... Attribute determination system 31 ... Analysis server 311 ... Control means 312 ... Communication means 313 ... history management means 314 ... attribute determination means 315 ... output means

Claims (6)

An attribute determination system for determining attributes indicating user characteristics,
Storage means for storing sample data of each medium indicating a user composition ratio for each attribute, and access history information of the user to the medium;
History management means for writing access history information to the storage means when a user terminal device accesses the medium; and
Attribute determination means for determining the attribute of the user from the access history information and sample data in the storage means;
An attribute determination system comprising: The attribute determination means determines an attribute by calculating a probability that the user has an attribute in the sample data.
The attribute determination system according to claim 1. Furthermore, the information determination means for collecting user information from the terminal device when the probability calculated by the attribute determination means is within a predetermined range,
The attribute determination system according to claim 2. And an output means for outputting media corresponding to the user attribute determined by the attribute determination means.
The attribute determination system according to any one of claims 1 to 3, characterized in that: An attribute determination method for determining an attribute indicating a feature of a user,
Preparing storage means for storing sample data of each medium indicating a user composition ratio for each attribute and access history information of the user to the medium,
When the user's terminal device accesses the medium, the access history information is written to the storage means,
Determining the attribute of the user from the access history information and sample data in the storage means;
An attribute determination method characterized by that. A computer program used in an attribute determination system for determining attributes indicating user characteristics,
A step of writing access history information when a user terminal device accesses the medium to a storage unit that stores sample data of each medium indicating a user composition ratio for each attribute and access history information of the user of the medium When,
Determining the attribute of the user from access history information and sample data in the storage means;
A computer program for causing a computer to execute.

Images