JP2002259440A - Advertisement transmission system and recording medium with advertisement transmission program recorded thereon - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an advertisement transmission system capable of transmitting an effective advertisement. SOLUTION: In the advertisement transmission system that is a server capable of transmitting/receiving data, through a network, between a user terminal that a user has and an advertisement information provider's server that an advertisement information provider has, the server has a means for receiving, from the advertisement information provider's server, the advertisement information to be distributed, a means for analyzing the advertisement attribute of the received advertisement information, a means for receiving the information on the user from the user terminal, a means for analyzing the attribute that the user has on the basis of the received information on the user, a means that compares the advertisement attribute with the information on the user and/or the attribute that the user has, and retrieves the user who is to be the distribution target for the advertisement information, and a means for distributing the advertisement information to the retrieved user.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an advertisement transmission system capable of performing effective advertisement transmission.

[0002]

2. Description of the Related Art Conventionally, advertisements on networks (in the present invention, advertisements indicate advertisements on the network, and advertisement media include banner advertisements, mail magazines, mail advertisements (opt-in mails, direct mails, etc.), etc. However, in the present embodiment, an e-mail advertisement is taken as an example. It should be noted that the present invention does not limit the present invention whatever the advertising medium. In order to eliminate waste, a destination is extracted in advance by some method, and an advertisement is transmitted to the extracted destination.

[0003] There are roughly two types of these advertisement transmission methods as follows. (1) The user consciously registers his / her own attribute and transmits an advertisement according to the attribute. (2) The profile of the user (behavior pattern on the network) is automatically analyzed without the user's awareness, the attribute is automatically registered, and the advertisement is transmitted accordingly.

The former advertisement transmission method is disclosed in Japanese Patent Laid-Open No. 9-163.
No. 352 and Japanese Patent Application Laid-Open No. Hei 9-251296, in addition to causing a burden on the user and using the user's self-report as it is, it cannot be confirmed whether the self-report is accurate or not. Also, it is difficult to transmit an advertisement suitable for a user who has not performed attribute registration. As a method for coping with this, there is an advertisement transmission method of determining an attribute of a user using an IP address (terminal position information) and transmitting an advertisement suitable for the user as disclosed in Japanese Patent Application Laid-Open No. 9-305518. However, with this method, it is hard to say that accurate attributes can be obtained at present when the network is diversified. In other words, although the IP address indicates the location of the server used by the user, it is frequently used by users in Japan who use the server in the United States and use the IP address, for example. This is because it cannot be said that a proper attribute has been obtained. Also, only a rough organization classification of the organization to which the user belongs can be obtained by this method, and it is difficult to obtain an attribute relating to the user's humanity such as the user's preference from Japanese Patent Application Laid-Open No. 9-305518.

The latter method of transmitting advertisements is disclosed in
As shown in JP-A-57157, a profile of a user is automatically analyzed without being conscious of the user, and attributes of the user are automatically acquired. However, in this method, although the data on the user attribute is always the latest, it is difficult to know the change in the user attribute. Therefore, it is natural that the attributes of the user change due to the passage of time and changes in the living environment (marriage, employment, childbirth, change of job, etc.), whereby the profile is also changed at any time (that is, having the latest profile). However, for example, it is difficult to retroactively know the attributes of a user ten years ago. In addition, profile data must be transmitted on the server to collect the user's profile, which increases network traffic and delays response to the user.

[0006] In addition, the attributes of the advertisement transmitted to the user by the conventional advertisement transmission method (regardless of the former or the latter) include:
Although the advertisement information provider or the manager (advertising sender) has subjectively determined this, this imposes a time and cost burden on the advertisement information provider or the administrator.

[0007]

In view of the above problems, the present inventors have devised an advertisement transmission system having the following features. (1) The user attributes that the user subjectively registers are extracted by using a statistical method to extract important elements, objectively re-evaluated from the entire population, and user attributes (such user attributes are factor attributes To verify). (2) By performing the operation of (1) regularly or irregularly, even if there is no new attribute registration or analysis of the user profile by the user, the user attribute change (change of the position that the user occupies in the whole or change in the past). At any point in time, the position occupied by the user can be grasped. (3) The attributes of the advertisement are automatically acquired by automatically analyzing the text of the advertisement to be transmitted to the user. (4) The user's response to the advertisement can be traced back in the past and the advertisement can be transmitted accordingly. That is, the user's reaction status from the past can be grasped. (5) The same user can be recognized through a plurality of sessions, and the user can be specified.

First, a user attribute which has been registered by the user subjectively is analyzed by a predetermined analysis means (in this embodiment, analysis is performed using quantification theory type III). Needless to say, other means such as regression analysis, multiple regression analysis, principal component analysis, discriminant analysis, factor analysis, canonical correlation analysis, quantification theory type I, quantification Theory Class II, Quantification Theory IV
Class, neural network analysis, decision tree analysis, cluster analysis, association rule analysis, and logistic regression analysis) to objectively reevaluate the attributes registered by the user. That is, by deriving the factor attribute, the user attribute can be grasped more accurately.

Second, by executing the first feature on a regular basis (it goes without saying that it may be irregular), for example, a user who was judged to be a user attribute having an “innovative” tendency 10 years ago However, it is possible to grasp the change of the factor attribute such as being determined to be “conservative” at present. That is, conventionally, once a user has registered a user attribute, the attribute does not change over time, and to change the user attribute, it is necessary to register a new user attribute or analyze a user profile. However, this feature is
The user only needs to register the user attribute once and does not need to analyze the user profile, and the user attribute in which the user is located in the population (that is, the factor attribute which is an objective user attribute) changes over time. To be able to grasp it. In addition, since there is no need to transmit / receive profile data to / from the server or to transmit a new user attribute, it is possible to reduce network traffic and reduce the load on the network.

Third, when the advertisement information provider transmits the advertisement information to the server (the advertisement transmission company has), the sentence is automatically analyzed even if the advertisement information provider or the advertisement transmission company does not register the attribute for the advertisement, Analyze the advertisement attribute to which the advertisement belongs. This eliminates artificial registration errors that occur during attribute registration of conventional advertisements, and makes it possible to select and register attributes more accurately. In addition, conventionally, the advertisement information provider or the advertisement transmission company has been imposed a time and cost burden, but it has also been made possible to reduce the burden.

Fourth, a user attribute database having a dynamic configuration enables the user attributes to be grasped retroactively. For example, a television is taken as an example. If the user is currently browsing television related advertisements frequently and the frequency of browsing is high compared to other advertisements, then the user is "currently" interested in television. That is, if an advertisement related to television is transmitted to the user, the efficiency of the advertisement is high. For that reason,
Many of the current advertising transmission systems offer their users
Focus on sending TV ads. However, if the user purchases a television at a certain point in time, the possibility of browsing after that even if an advertisement relating to the television is transmitted is extremely reduced (that is, the browsing rate is reduced). Then, the current advertisement transmission system stops transmitting the television advertisement to the user based on the decrease in the browsing rate of the advertisement. Assuming the average TV life cycle is 5 years, re-sending TV ads "5 years later" from the point of purchase (when the ad browsing rate decreases)
The browsing rate for the ad will improve. But,
The current advertisement transmission system has the latest user attribute, but is difficult because it cannot be traced back to the past.
The fourth covers this point. Thus, for example, if an attribute is for an object, it is possible to transmit an advertisement by referring to a past user attribute in consideration of its life cycle and the like. That is, at the time of transmitting an advertisement, it is possible to refer to not only the latest user attribute but also the past user attributes, thereby transmitting the advertisement, improving the advertisement browsing rate, and performing effective advertisement transmission.

Fifth, conventionally, there is a method of collecting and analyzing user attributes when clicking an advertisement on the web, using a cookie, an IP address, or the like, but this method is the same for each session. Since different IDs are assigned to users, it is difficult to accumulate information on the same user through a plurality of sessions. However, in the present invention, a user to be a transmission destination is specified, and a response is received from the user by the server, whereby the user can be specified and its analysis can be performed.

[0013]

According to the first aspect of the present invention, an advertisement is a server which can transmit and receive data between a user terminal of a user and an advertisement information provider server of an advertisement information provider via a network. In the transmission system, the server receives advertisement information to be distributed from the advertisement information provider server, analyzes an advertisement attribute of the received advertisement information, and transmits information about the user from the user terminal. Receiving means, means for analyzing attributes of the user based on the received information on the user, and comparing the advertisement information with the information on the user and / or the attributes of the user. Means for searching for the user to be distributed, and means for distributing the advertisement information to the searched user. That is an advertisement transmission system.

According to a second aspect of the present invention, the means for analyzing the attribute of the advertisement information analyzes a sentence of the advertisement information and extracts a frequently-used word extracted as a result of the sentence analysis and a keyword group to which the frequently-applied word belongs. An advertisement transmission system that determines the advertisement attribute by making a comparison.

According to a third aspect of the present invention, the keyword group includes:
This is an advertisement transmission system including at least one word having similarity or similarity to each other.

The invention according to claim 4, wherein the similarity and the similarity are provided by giving a numerical value to the word and calculating a correlation coefficient or a correlation between the words based on the numerical value. It is.

An invention according to claim 5 is an advertisement transmission system using the keyword group as the advertisement attribute.

The invention according to claim 6 is an advertisement transmitting system in which the means for analyzing the attribute of the user analyzes using a statistical method.

According to a seventh aspect of the present invention, the server receives an advertisement attribute of advertisement information distributed to the user and an advertisement attribute of advertisement information to which the user has responded, and periodically and / or irregularly. This is an advertisement transmission system that further includes a unit that stores the data on a per-division basis.

The invention according to claim 8 is an advertisement transmitting system for analyzing by calculating factor attributes of the user by using quantification theory type III as the statistical method.

According to a ninth aspect of the present invention, the statistical methods include regression analysis, multiple regression analysis, principal component analysis, discriminant analysis, factor analysis, canonical correlation analysis, quantification theory type I, and quantification theory II. This is an advertisement transmission system for analyzing by calculating the factor attribute of the user by using any one of class, quantification theory class IV, neural network analysis, decision tree analysis, cluster analysis, association rule analysis, and logistic regression analysis.

According to a tenth aspect of the present invention, the means for searching for a user to whom the advertisement information is to be distributed includes at least the advertisement attribute, information about the user and / or a factor attribute of the user, and / or distribution in the past. An advertisement transmission system for performing a search based on the number of reactions of the user to an advertisement attribute included in the advertisement information obtained.

According to an eleventh aspect of the present invention, there is stored an advertisement transmission program stored in a server capable of transmitting and receiving data via a network between a user terminal of a user and an advertisement information provider server of an advertisement information provider. In the medium, the advertisement transmission program receives advertisement information to be distributed from the advertisement information provider server, analyzes an advertisement attribute of the received advertisement information, receives information about the user from the user terminal, Analyzing the attribute of the user based on the received information on the user, comparing the advertisement attribute with the information on the user and / or the attribute of the user, and distributing the advertisement information to the user; And an advertisement transmission program for delivering the advertisement information to the searched user. It is a medium.

[0024]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described in detail with reference to the drawings. FIG. 1 is an example of a system configuration diagram of the present invention. In the present embodiment, the advertisement transmission system 1 composed of the server 2 uses a personal computer or a PDA (Per
personal digital assistant), an advertisement information provider server 4 operated by an advertisement information provider, and the Internet or a LAN (Local).
It is connected by a network 5 typified by Area Network).

The server 2 transmits an advertisement provided by the advertisement information provider to the user terminal 3 according to the attribute of the user. The server 2 includes a user attribute registration unit 6, a user attribute database 7, a user attribute analysis unit 8, a user search unit 9, a dynamic user attribute database 10, an advertisement attribute database 11, an advertisement attribute analysis unit 12, an advertisement information reception unit 13,
Keyword index database 14, advertisement database 1
5. It has distribution means 16.

The user attribute registration means 6 is a means for transmitting a user attribute input screen (a screen for inputting a questionnaire) to the user terminal 3 and registering the questionnaire result information in the user attribute database 7.

The user attribute database 7 stores basic information such as the name, gender, age, occupation, and field of interest of each user and attribute information based on a questionnaire in a database.

The user attribute analysis means 8 patterns attribute information such as sex, age, field of interest, and personality obtained by a questionnaire at the time of user registration, and derives a factor attribute which is an objective attribute of the user. Means.

The user search means 9 is a means for searching for a destination of an advertisement to be transmitted according to a factor attribute (or a user attribute) or an advertisement attribute. The user search means 9 includes a user attribute database 7, a dynamic user attribute database 10, and an advertisement attribute database. 11 is a means for searching for necessary information. When searching for a factor attribute, the factor attribute is derived according to an already stored answer pattern.

The dynamic user attribute database 10 is a database that stores the number of transmitted advertisement information and the number of clicked advertisement information in an advertisement information attribute (advertisement attribute) column defined for each user in advance. . This data is divided at regular intervals (every day, every month, etc.) over time and stored in a database. Ad attributes are categorized in detail, and the number of clicks (reactions) of ad information with specific attributes within a certain period of time is
It measures the user's interest in advertising information and the degree of interest. Therefore, it is determined that a user who has a higher response rate to an advertisement having a specific attribute within a certain period of time has a higher interest in the advertisement information having the corresponding attribute. FIG. 7 shows the structure of the dynamic user attribute database 10.

The advertisement attribute database 11 is a database that stores a group of keywords used when determining an advertisement attribute of an advertisement provided by an advertisement information provider. That is, it is a database in which the keyword group of the advertisement to be transmitted is registered and stored as the advertisement attribute. FIG. 8 shows the advertisement attribute database 11.
An example of the structure of FIG.

The advertisement attribute analysis unit 12 is a unit that analyzes the advertisement attribute of the advertisement provided by the advertisement information provider and registers the advertisement attribute in the advertisement attribute database 11. The method of analysis is to perform sentence analysis on the provided ads,
From the text of the advertisement (advertisement information), several to dozens of keywords are extracted. The keyword index database 14 determines where each of these keywords belongs to the keyword group registered in the advertisement attribute database 11.
(See below). Since the corresponding keyword group indicates the advertisement attribute of the advertisement, this data is transmitted to the advertisement attribute database 11 and stored.

The advertisement information receiving means 13 is a means for receiving the advertisement information transmitted by the advertisement information provider server 4 and transmitting the received advertisement information to the advertisement attribute analysis means 12. After the advertisement is transmitted, the advertisement database 15
To send the advertisement information.

The keyword index database 14 is a database in which information as to which keyword group a keyword (word) corresponds to is registered. The structure of the keyword group is calculated as follows. A plurality of words (groups) determined to have a high correlation by a known method are formed by a known method with respect to the words extracted as a result of the sentence analysis of the advertisement information stored in the advertisement database 15 by the advertisement attribute analysis unit 12. The group of words is referred to as a keyword group. FIG. 15 shows an image diagram of the keyword group. The database storing the correspondence between keywords and keyword groups is the keyword index database 14. FIG. 9 shows an example of the structure of the keyword index database 14.

The new advertisement information is transferred to the advertisement information provider server 4
When the advertisement information receiving means 13 receives the advertisement information from the advertisement information analyzing means 12, the advertisement attribute analyzing means 12 analyzes the sentence of the advertisement information received from the advertisement information receiving means 13, and as a result, when "outside car" is applicable as a frequent word, " A keyword group of the word "outside car" is searched from the keyword index database 14,
In this case, since “outside car” and “car” correspond to each other, “car” is applicable as a keyword group, and this advertisement has an advertisement attribute of “car”. In the present embodiment, the keywords belong to one keyword group, but it goes without saying that the keywords may belong to a plurality of keyword groups.

The advertisement database 15 stores the advertisement information provided by the advertisement information provider in a database.
That is, it is a database in which the advertisement information transmitted to the user is stored.

The distribution unit 16 transmits an advertisement to the user terminal 3 of the transmission destination of the transmitted advertisement searched by the user search unit 9 based on the advertisement attribute to be transmitted analyzed by the advertisement attribute analysis unit 12. It is a means to do. The transmission result of the transmitted user is transmitted to the dynamic user attribute database 10 and updated.

[0038]

FIG. 1 shows an example of the process flow of the present invention.
This will be described in detail with reference to an example of a system configuration diagram, flowchart diagrams of FIGS. 2 to 6, conceptual diagrams of FIGS. 7 to 21, and an example of a screen image.

The user operates the U of the server 2 from the user terminal 3.
RL (Uniform Resource Locato
The server 2 is accessed by inputting r). If the user uses the advertisement transmission system 1 for the first time, user registration is performed. At the time of user registration, the user answers a detailed questionnaire on user attributes such as age, family structure, taste characteristics, and consumption tendency (S100). The user attribute registration unit 6 transmits a user attribute input screen (questionnaire) to the user terminal 3. The contents of the questionnaire include gender, age, address, marriage status, family structure, occupation, business type, Internet use environment, Internet use experience and purpose, owned car type, hobby, interest, interest, income, and the like. FIG. 10 shows an example of a user attribute input screen.
Shown in FIG. 20 shows an example of a user attribute to be queried on the user attribute input screen.

The new user attribute information input from the user terminal 3 through the user attribute input screen in S100 and received by the user attribute registration means 6 of the server 2 is (S11
0), received by the user attribute registration means 6 of the server 2, registered and stored in the user attribute database 7 (S12).
0). The new user information registered in the user attribute database 7 is transmitted to the user attribute analysis means 8 and the user attribute analysis means 8 determines a factor attribute which is an objective evaluation of the user attribute. It may be verified and corrected whether the user attribute information registered voluntarily by the user matches the attribute in the population. At the same time, the pattern of the entire factor attribute is analyzed regularly or irregularly to classify it into factor attributes occupied by users in the population (S40).
0).

As the analysis method in S400, there are a plurality of known analysis methods such as multivariate analysis and quantification theory. In the present embodiment, quantification theory type III is used as the method. It goes without saying that the method may be used. Examples of other methods include regression analysis, multiple regression analysis, principal component analysis, discriminant analysis, factor analysis, canonical correlation analysis, quantification theory type I, quantification theory type II, quantification theory type IV,
There are a neural network analysis, a decision tree analysis, a cluster analysis, an association rule analysis, a logistic regression analysis and the like.

The process of calculating a factor attribute of a user using quantification theory type III will be described below. As stored in the user attribute database 7 in step S120, the user attribute information indicates each genre, which is a question item on the user attribute input screen such as "personality" and "consumption behavior", and the answer and tendency to the genres. The information is classified into categories and stored in the user attribute database 7. FIG. 11 shows an example of the structure of the user attribute database 7. In the present embodiment, the “personality” genre is taken as an example, but the following work is performed in each of the other genres. The “personality” genre has “innovative”, “logical”, “conservative” categories and the like, and the result of self-diagnosis at the time of the questionnaire is converted into a number of 1 or 0 as presence or absence of each category, and the It is stored in the attribute database 7.

In the “personality” genre, if there are i categories including the “innovative” category, and there are k users, the reaction of the k-th user with respect to the i-th category (relevant Is the variable Xi (k)
And the variable Xi (k) is either 1 or 0.

The user attribute analysis means 8 extracts all (or a part of) the user attribute information stored in the user attribute database 7 (S410). Category score in the population used in the quantification theory class III from the extracted user attribute information (measured based on the statistical distance between each category against any meaningful standard (factor axis)) , A positive number indicates a positive distance relationship with the axis (positive for the category score), and a negative number indicates the opposite relationship. (Measured up to) (S1)
20). FIG. 12 shows an example of the calculated category score. Therefore, the numerical value of the category score can be expressed as a variable Ai (j).

If the category score has been calculated in S420, the meaning of each factor axis is interpreted while referring to the category and the relationship between the category score and the factor axis (S430). . That is, the interpretation is performed by showing the category scores for each factor axis as a graph as shown in FIG. FIG. 13 shows an example of the first factor axis. In the example shown in FIG. 11, the conservative, innovative, and logical categories are in opposition, so the positive direction of the axis is "traditional respectful and cautious," and the negative direction of the axis is "new." The positive type "is interpreted. That is, two interpretations (positive interpretation, negative interpretation) are performed for each factor axis, and if neither of the positive or negative interpretations applies, a neutral interpretation is performed, and the total for one genre is calculated. 2j + 1 (2 × j + 1) interpretations are performed, and the user has at least one of these factor attributes.

If the interpretation has been made for all the factor axes in S430, a sample score for each factor axis of each user is calculated based on the calculated category scores (S440). The sample score expresses the user's characteristics (how to stand out) on each factor axis, and is calculated based on Equation 1 derived from Quantification Theory III.

## EQU1 ## Sample score on the j-axis of the k-th person = Σ (X
i (k) × Ai (j)) ÷ ΣXi (k)

The number most prominently expressed for the sample score in each factor axis of any user is the largest value among the absolute values of each sample score.
According to the numerical value, the factor attribute of the user can be determined. That is, the factor axis to which the factor attribute of the user belongs is determined by the equation shown in Expression 2.

## EQU2 ## Factor axis to which the user belongs = max (| Σ (Xi (k)
× Ai (j)) ÷ ΣXi (k) |)

When the factor axis to which the user belongs in Equation 2 is determined, the factor attribute to which the user belongs is determined by whether or not the sample score in Equation 1 is a positive number. From the calculation result of Equation 2, if the factor axis to which the user belongs is the first factor axis and the sample score calculated from Equation 1 is a positive number, it is a "traditional respectful cautious type". In particular, the factor attribute of the user is determined to be "positive type", or if neither of the above is true, "neutral type". In the present embodiment, the factor attribute that the user has for one genre is the only one that is most prominent from each factor axis. It may have an attribute. In this case, the attribute for each factor axis may be derived based on the sample score calculated from Equation 1, and the factor attribute of the user may be derived from a plurality of factor axes.

By using this, the user can execute S10
For example, from the answer pattern input to the user attribute input screen at 0, for example, "innovative" =
1, the user who inputs "positive" = 0 and "conservative" = 0 can set the factor attribute of the user as "positive for new type". That is, from the answer pattern input on the user attribute input screen, the user can determine which factor attribute corresponds to s (2j + 1) ways (however, the number of genres is s
To be selected).

Therefore, if the s (2j + 1) answer patterns are analyzed regularly or irregularly using the user attribute database 7, the attributes of the users in the population can be verified and the factor attributes can be set. Becomes possible. It is clear that by performing this analysis regularly or irregularly, the factor attribute possessed by the user changes as the population changes. That is, the user changes the factor attribute, which is an objective evaluation of the user, with only one input, and can change the position occupied by the user in the population. After the analysis from S400 to S440, this answer pattern (pattern for each genre of factor attribute) is transmitted from the user attribute analyzing means 8 to the user searching means 9 and the user searching means 9 receives s (2j + 1) different interpretations (factor attributes). An answer pattern for the answer is stored and registered (S45).
0).

When the new user attribute information inputted from the user terminal 3 is received by the user attribute registering means 6, the new user attribute information is transmitted to the user attribute analyzing means 8, and the answer pattern of the new user attribute information and S450 In step S130, a factor attribute of the new user is determined by collating with the answer pattern stored in the user search means 9. In this case, it is needless to say that the matching may be performed only for a specific genre, instead of matching the answer patterns of all the new user attribute information.

Next, the process of automatic classification of the advertisement attribute is shown in FIG.
This will be described with reference to the flowchart shown in FIG.
0). The advertisement database 15 stores the already transmitted advertisement information. This advertisement database 15 is searched by the advertisement attribute analysis means 12, and frequent keywords are extracted from frequently stored words from the stored advertisement information (for example, 50,000 frequent keywords) (S510). Next, positive or negative similarity / similarity is compared for the extracted 50,000 frequently-used keywords. The comparison of similarity / similarity is realized by comparing the correlation coefficient or correlation between frequently appearing words extracted from the same advertisement information. In some cases, a known context analysis method may be added to compare the similarity / similarity. As described above, the frequent words are converged to the corresponding keyword group (S520), and the frequent words are stored in the advertisement attribute database 11 (S530). That is,
In the advertisement attribute database, the keyword group is held as advertisement attributes. Further, the correspondence between the keyword group and frequently appearing words belonging to the keyword group is registered and stored in the keyword index database 14.

On the other hand, when the advertisement information provider attempts to transmit its own advertisement through the advertisement transmission system 1, the advertisement (advertisement information) transmitted from the advertisement information provider server 4 is transmitted to the server 2 and the transmission request is transmitted. I do. FIG. 14 illustrates an example of the transmitted advertisement.

When the transmission request and the advertisement information from the advertisement information provider server 4 are received by the advertisement information receiving means 13 of the server 2 (S140), the advertisement information is transmitted to the advertisement attribute analysis means 1
Send to 2. When the advertisement attribute analysis unit 12 receives the advertisement information, the advertisement attribute analysis unit 12 executes a sentence analysis of the advertisement information and extracts a keyword included in the advertisement information (S
150).

The keyword index database 14 determines which of the keyword groups stored in the advertisement attribute database 11 the keyword extracted in S150 belongs to.
To determine. Since the corresponding keyword group is the attribute of the advertisement information, it is stored in the advertisement attribute database, and the attribute of the advertisement information is automatically classified (S16).
0).

After the advertisement attributes are automatically classified in step S160, the server 2 selects whether to automatically extract the advertisement information destination or manually search the advertisement information destination server in the advertisement information provider server 4. Send the screen. FIG. 16 shows an example of the destination extraction selection screen. The advertisement information provider selects whether to automatically extract or manually search for the destination of the advertisement information from the destination extraction selection screen received by the advertisement information provider server 4 (S170).

On the transmission destination extraction selection screen of S170,
If the advertising provider chooses to extract automatically,
Using the advertisement attribute information classified in S160 as a search key, a corresponding factor attribute and / or user attribute is searched from the user attribute database 7 and / or the dynamic user attribute database 10 (S180), and extracted from each database. By doing so (S190), users to be delivered are extracted. At this time, it may be based on a factor attribute which is an objective evaluation of the user, or may be based on a user attribute which is a voluntary registration of the user. Also, in consideration of the life cycle of the product corresponding to the keyword included in the advertisement attribute, the search may be automatically performed in consideration of the user's reaction period (elapsed period from the time when the advertisement information was previously delivered to the user). good.

The number of relevant users extracted in S450 is transmitted to the advertisement information provider server 4 and displayed (S20).
0). Generally, the cost varies depending on the number of users who transmit advertisements, and thus the advertisement information provider may be able to determine the advertisement attribute based on the number of users. The advertisement information determined as described above is transmitted from the user search means 9 to the distribution means 16 and transmitted from the distribution means 16 to the user terminal 3 of the corresponding user (S210).

The distribution unit 16 transmits the advertisement attribute of the transmitted advertisement information to the dynamic user attribute database 10, and registers and stores the transmitted advertisement attribute for each user (S22).
0). At the same time, the advertisement attribute analysis means 12 stores the new advertisement information transmitted to the advertisement database 15.

The user terminal 3 that has received the advertisement information transmitted in S220 browses the advertisement information (S23).
0). When the user clicks the advertisement information and shows some reaction, the user terminal 3 sends information to the server 2 for specifying the user (for example, a user ID) and information for specifying the advertisement information (for example, the advertisement ID Is transmitted) and acquired by the server 2 (S240). When the distribution means 16 of the server 2 receives the user ID and the advertisement ID, the server jumps to the advertisement information provider server 4 and displays a more detailed advertisement on the user terminal 3. That is, the web page of the URL is displayed on the user terminal 3.

Further, the distribution means 16 transmits the advertisement ID received in S240 to the advertisement attribute analysis means 12, and the advertisement attribute analysis means transmits the advertisement ID to the advertisement attribute database 1 based on the advertisement ID.
1 is acquired (S250).
The advertisement attribute and the user ID acquired in S250 are transmitted from the distribution unit 16 to the dynamic user attribute database 10, and information indicating that the user clicked (reacted) on the advertisement information is registered and stored in the dynamic user attribute database 10. Is performed (S260). That is, the dynamic user attribute database 10
Is registered and stored as "1" (transmission is present) for each advertisement attribute of the transmitted advertisement, and the advertisement attribute is registered and stored as transmitted information.

When the advertisement information is to be transmitted from now on and the user wants to know the reaction to the past transmission result, the server 2 requests and displays the transmission result search screen. FIG. 21 shows an example of the transmission result search screen. The advertisement information provider is the advertisement information provider server 4
The server 2 transmits the selection information such as the period to be extracted from the information, and the server 2 receives the selection information. The server 2 extracts, from the dynamic user attribute database 10, the number of transmissions and the number of reactions of the advertisement information corresponding to the attribute specified within a certain period based on the selection information,
The click number confirmation screen in which these are described is transmitted to the advertisement information provider server 4 and displayed (S270). FIG. 17 shows an example of the reaction number confirmation screen.

When the user wants to know the reaction of the advertisement information transmitted by the advertisement information provider, a request for confirming the reaction result is made from a screen (reaction result confirmation screen) (not shown) for confirming the reaction result of the corresponding advertisement information. Transmit and the server 2 receives the confirmation request. The server 2 extracts the number of transmissions and the number of reactions of the advertisement information corresponding to the advertisement from the dynamic user attribute database 10 based on the confirmation request, and displays an advertisement reaction confirmation screen (not shown) in which the information is described. The data is transmitted to the server 4 and displayed. At this time, the advertisement may be transmitted without the request of the advertisement information provider, and the server 2 may automatically transmit the advertisement reaction confirmation screen to the advertisement information provider server 4 and display it after a certain period of time.

In the transmission destination extraction selection screen of S170,
If the advertiser chooses to do it manually,
It is determined whether or not the search is a natural sentence (S28).
0). When the search using the natural sentence is selected in S280, the natural sentence input screen is transmitted from the server 2 to the advertisement information provider server 4. The natural sentence input screen is a screen on which a user search can be executed using a natural sentence, and an example thereof is shown in FIG.

The advertisement information provider inputs a search sentence (for example, “a person who is sensitive to fashion and brand and is interested in sports cars”) from the advertisement information provider server 4 in the input field of the natural sentence input screen (S290). . The entered search text is sent from the advertisement information provider server 4 to the user search means 9 of the server 2.
Send to Upon receiving the search sentence, the user search means 9 executes a sentence analysis of the search sentence and extracts a keyword included in the search sentence (S300) (for example, "fashion"
"Brand" and "sports car" are extracted.)

The keyword extracted in S300 is inquired of the keyword index database 14, and a search is performed to determine to which keyword group the extracted keyword belongs. That is, in this example, the keyword “fashion” is a keyword group “fashion” and a keyword “brand”
Corresponds to the keyword group “fashion”, and the keyword “sports car” corresponds to the keyword group “car”. In addition, the advertisement information provider inputs and selects a reaction period (that is, when to extract data from after) on the natural sentence input screen from the advertisement information provider server 4 (S31).
0). The dynamic user attribute database 10 is searched from the reaction period and the keyword group selected in S310,
The total number of reactions for each user is calculated and scored. The scored users are extracted from the dynamic user attribute database 10 (S320), and the processes from S200 to S270 are executed. It is not essential whether or not the reaction period is selected in S310, but by making the selection, effective advertisement transmission can be performed in consideration of the life cycle and the like.

When the search based on the natural sentence is not executed in S280, the attribute search screen is transmitted from the server 2 to the advertisement information provider server 4. The attribute search screen is a screen for extracting a user as a transmission destination based on a reaction period, the number of reactions, a reaction attribute, and the like, and transmitting advertisement information to the user. FIG. 19 shows an example of the attribute search screen.

The advertisement information provider selects the information at an arbitrary time in the past through the attribute search screen (selection of the reaction period) (S330), or targets the user who has responded. (S340), and what kind of advertisement attribute the user who responded to is selected (S350). Based on the information selected in S330 to S350, the user corresponding to the search is extracted from the dynamic user attribute database 10, and the processes in S200 to S270 are executed.

By selecting the reaction period in S310 or S330, it becomes possible to perform the advertisement transmission retroactively to the user attributes according to the attribute, which was difficult in the past. In other words, if it is desired to transmit advertisement information on a certain product, a user whose reaction was remarkable at a point in time when the user retrospects for a period corresponding to the life cycle of the product is extracted, and if the advertisement information can be transmitted to the user, As a result, the advertisement information can be transmitted more efficiently.

In carrying out the present invention, a storage medium storing a software program for realizing the functions of the present embodiment is supplied to the system, and the computer of the system reads and executes the program stored in the storage medium. It is natural to be realized by. Further, in this embodiment, the number of servers is one, but a plurality of servers or terminals may be provided separately.

In this case, the program itself read from the storage medium implements the functions of the above-described embodiment, and the recording medium storing the program naturally constitutes the present invention.

As a storage medium for supplying the program code, for example, a floppy (registered trademark) disk,
A hard disk, an optical disk, a magneto-optical disk, a magnetic tape, a nonvolatile memory card, or the like can be used.

When the computer executes the readout program, not only the functions of the above-described embodiment are realized, but also the operating system running on the computer based on the instructions of the program. It goes without saying that a part or all of the processing is performed and the functions of the above-described embodiments are realized by the processing.

Further, after the program read from the storage medium is written in a nonvolatile or volatile storage means provided in a function expansion board inserted in the computer or a function expansion unit connected to the computer, the program is read out. Based on the instruction, the arithmetic processing device or the like provided in the function expansion board or the function expansion unit may perform part or all of the actual processing, and the processing may realize the functions of the above-described embodiments. It is.

[0075]

According to the present invention, (1) it is possible to objectively evaluate a position in a population based on a user attribute input voluntarily by a user. (2) By using the concept of a factor attribute, a user can input a user attribute only once and automatically correct how the subsequent user attributes change in the population. It has become possible. (3) By automatically analyzing attributes included in the advertisement information, human errors can be prevented, and accurate advertisement attributes can be added, and more effective advertisement transmission can be performed. (4) By being able to retroactively analyze the reaction of the user in the past, it has become possible to produce more effective advertising effects, such as transmitting advertisements in consideration of the product life cycle and the like. (5) By adding an objective evaluation to the process, the destination can be more accurately targeted than the conventional advertisement transmission system, and time and cost can be reduced. (6) The same user can be recognized through a plurality of sessions, and the user can be specified.

[Brief description of the drawings]

FIG. 1 is an example of a system configuration diagram showing an example of a system configuration of the present invention.

FIG. 2 is a first sheet of a flowchart showing an example of the process of the present invention.

FIG. 3 is a second flowchart illustrating an example of the process of the present invention.

FIG. 4 is a third flow chart showing an example of the process of the present invention.

FIG. 5 is a flowchart illustrating an example of a factor attribute pattern process.

FIG. 6 is a flowchart illustrating an example of an advertisement keyword classification process.

FIG. 7 is a conceptual diagram showing a structure of a dynamic attribute database.

FIG. 8 is a conceptual diagram showing the structure of an advertisement attribute database.

FIG. 9 is a conceptual diagram showing a structure of a keyword index database.

FIG. 10 is a diagram showing an example of a user attribute input screen.

FIG. 11 is a conceptual diagram showing a structure of a user attribute database.

FIG. 12 is a conceptual diagram illustrating calculation of a category score.

FIG. 13 is a conceptual diagram showing interpretation of a factor axis.

FIG. 14 is a diagram showing an example of an advertisement to be transmitted.

FIG. 15 is a conceptual diagram showing a relationship between a keyword and a keyword group.

FIG. 16 is a diagram illustrating an example of a destination extraction selection screen.

FIG. 17 is a diagram showing an example of a reaction number confirmation screen.

FIG. 18 is a diagram showing an example of a natural sentence input screen.

FIG. 19 is a diagram showing an example of an attribute search screen.

FIG. 20 is a diagram showing an example of a list of user attributes to be asked on the user attribute input screen.

FIG. 21 is a diagram illustrating an example of a transmission result search screen.

[Explanation of symbols]

 1: Advertisement transmission system 2: Server 3: User terminal 4: Advertisement information provider server 5: Network 6: User attribute registration means 7: User attribute database 8: User attribute analysis means 9: User search means 10: Dynamic user attribute Database 11: Advertising attribute database 12: Advertising attribute analyzing means 13: Advertising information receiving means 14: Keyword index database 15: Advertising database 16: Distribution means

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ Identification symbol FI Theme coat ゛ (Reference) G06F 17/60 ZEC G06F 17/60 ZEC (72) Inventor Ryota Furuya 27-2 Nihonbashi Hakozakicho, Chuo-ku, Tokyo F-term in Cyber Brains Inc. (reference) 5B075 KK07 KK34 KK40 ND20 PQ02 PR08 QM10 UU40

Claims (11)

[Claims] 1. An advertisement transmitting system which is a server capable of transmitting and receiving data via a network between a user terminal of a user and an advertisement information provider server of an advertisement information provider. Means for receiving advertisement information to be provided from the advertisement information provider server, means for analyzing advertisement attributes of the received advertisement information, means for receiving information about the user from the user terminal,
Means for analyzing an attribute of the user based on the received information on the user, and a target for distributing the advertisement information by comparing the advertisement attribute with the information on the user and / or the attribute of the user An advertisement transmission system, comprising: means for searching for the user; and means for delivering the advertisement information to the searched user. 2. The means for analyzing the attribute of the advertisement information, wherein the frequent word extracted as a result of the sentence analysis by analyzing the sentence of the advertisement information is compared with a keyword group to which the frequent word belongs. The advertisement transmission system according to claim 1, wherein the advertisement attribute is determined. 3. The advertisement transmitting system according to claim 2, wherein the keyword group is composed of at least one word having similarity or similarity with each other. 4. The method according to claim 3, wherein the similarity and the similarity are obtained by assigning a numerical value to the word and calculating a correlation coefficient or a correlation between the words based on the numerical value. The advertisement transmission system according to 1. 5. The advertisement transmitting system according to claim 1, wherein the keyword group is used as the advertisement attribute. 6. The advertisement transmitting system according to claim 1, wherein the means for analyzing the attribute of the user analyzes using a statistical method. 7. The server receives an advertisement attribute of advertisement information distributed to the user and an advertisement attribute of advertisement information to which the user has responded, and stores the advertisement attribute at regular intervals and / or at irregular intervals. The advertisement transmitting system according to claim 1, further comprising a unit. 8. The quantification theory II as the statistical method.
The advertisement transmission system according to any one of claims 1 to 7, wherein analysis is performed by calculating a factor attribute of the user using Class I. 9. The statistical methods include regression analysis, multiple regression analysis, principal component analysis, discriminant analysis, factor analysis, canonical correlation analysis, quantification theory type I, quantification theory type II, quantification theory The analysis is performed by calculating a factor attribute of the user using any one of Class IV, neural network analysis, decision tree analysis, cluster analysis, association rule analysis, and logistic regression analysis. 8. The advertisement transmission system according to any one of 7. 10. The means for searching for a user to whom the advertisement information is to be distributed has at least the advertisement attribute, the information on the user and / or the factor attribute of the user, and / or the advertisement information distributed in the past. The advertisement transmission system according to any one of claims 1 to 9, wherein the search is performed based on the number of reactions of the user to advertisement attributes. 11. A storage medium storing an advertisement transmission program of a server capable of transmitting and receiving data via a network between a user terminal of a user and an advertisement information provider server of an advertisement information provider, The advertisement transmission program receives the advertisement information to be distributed from the advertisement information provider server, analyzes an advertisement attribute of the received advertisement information, receives information about the user from the user terminal, and receives the information about the received user. Analyzing the attribute of the user based on information, searching for the user to whom the advertisement information is to be distributed by comparing the advertisement attribute with information on the user and / or the attribute of the user, A storage storing an advertisement transmission program, wherein the advertisement information is distributed to a searched user. Body.