JP2012208562A - User selection apparatus, user selection method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To spread a campaign effectively.SOLUTION: A user selection apparatus 1 comprises: a candidate user selection part to select candidate users with interest-and-idea matching with a campaign from among users; a calculation part to calculate the number of acquaintances in a group of the candidate users for every candidate user; a generation part to generate distribution of the calculated number of acquaintances of each candidate user; and a target user selection part to select the candidate of the campaign based on the generated distribution.

Description

  The present invention relates to a user selection device, a user selection method, and a program for selecting a campaign target person.

In corporate marketing strategies, etc., it is considered that by identifying people with many acquaintances and conducting campaigns intensively, aiming for the ripple effect in word of mouth and the effect of reducing advertising costs when deploying services .
In Patent Document 1, it is assumed that a person having many acquaintances is a user having a high degree of importance, and the acquaintance number k of the user and the acquaintance number k in the social network topology of the campaign object person are given to the campaign object user. There is disclosed a method for calculating the degree of importance of a user based on the number of other users possessed.

JP 2010-204842 A

  However, in the technique described in Patent Document 1, a social network is generated based on the traffic volume of a mobile phone and the number of SNS (Social Network Service) users, and does not take into account personal preferences. For this reason, even if a campaign is conducted for a highly important user in this social network, if the highly important user is not interested in the campaign, the topic of the campaign will appear in the conversation between the users. There is a problem that the campaign cannot be effectively propagated.

  The present invention has been made in view of the above points, and an object of the present invention is to provide a user selection device, a user selection method, and a program capable of effectively spreading a campaign.

  The present invention has been made in order to solve the above-described problem, and a candidate user selection unit that selects a candidate user having a hobby and a taste that matches a campaign from among users, and for each candidate user, the candidate user's A calculation unit that calculates the number of people in the group, a generation unit that generates a distribution of the number of people of each candidate user calculated by the calculation unit, and a target person of the campaign based on the distribution generated by the generation unit A target user selection unit for selecting a user selection device.

  The user selection device may further include a first storage unit that stores a website browsing history for each user, and the candidate user selection unit selects a user who browsed a website that matches the campaign as the candidate. It is characterized by selecting as a user.

  The user selection device may further include a second storage unit that stores a call history or a mail transmission / reception history for each user, and the calculation unit is configured to transmit or receive a call or a mail with the first candidate user. The candidate user is an acquaintance of the first candidate user.

  The user selection device may further include a third storage unit that stores a friend of each user on the website, and the calculation unit selects a first candidate user and another candidate user who is a friend on the website. The acquaintance of the first candidate user is used.

  In addition, the present invention provides a step in which a user selection device selects a candidate user having a hobby and a taste that matches a campaign from among users, and the user selection device knows in the set of candidate users for each candidate user. A step of calculating the number of people, a step in which the user selection device generates a distribution of the calculated number of known users of each candidate user, and the user selection device selects a target person of the campaign based on the generated distribution. And a user selecting method.

  The present invention also includes a step of selecting candidate users having hobbies and preferences that match a campaign from among users, a step of calculating the number of people in the set of candidate users for each of the candidate users, and each of the calculated candidates It is a program for causing a computer to execute a step of generating a distribution of the number of acquaintances of users and a step of selecting a target person of the campaign based on the generated distribution.

  According to the present invention, candidate users having hobbies and preferences that match the campaign are selected, the number of people of each candidate user in the selected set of candidate users is calculated, and the campaign target person is selected based on the calculated distribution of the number of people. You can effectively spread the campaign to make a choice.

It is the schematic which shows the structure of the social network production | generation system by one Embodiment of this invention. It is a block diagram which shows the function structure of the user selection apparatus by this embodiment. It is a flowchart which shows the procedure of the campaign object user selection process by this embodiment.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings.
FIG. 1 is a schematic diagram illustrating a configuration of a social network generation system according to an embodiment of the present invention.
The social network generation system is a system for selecting a user to be campaigned, and includes a user selection device 1, a communication history database server 2, a call / mail history database server 3, and a terminal of each user. The apparatus 4 is comprised. A social network is a network topology in which a person (user) is a node and a relationship for exchanging information is a link.
The user selection device 1 and the communication history database server 2, and the user selection device 1 and the call / mail history database server 3 are connected by a LAN (Local Area Network) or the like. Further, the user selection device 1, the communication history database server 2, the call / mail history database server 3 and the terminal device 4 of each user are connected via a network such as the Internet.

  The terminal device 4 of each user is a terminal such as a mobile phone, a smartphone, a PHS (Personal Handy-phone System), or a personal computer.

  The communication history database server 2 is a server device that includes a database that stores a communication history for each user. The communication history is a history of information related to communication of each terminal device 4. Information related to communication includes browsing history of websites, purchase history of applications of the terminal device 4 (hereinafter referred to as mobile applications), purchase history of websites that carry out mail order, websites (Twitter (registered trademark) and blogs). SNS, etc.), contents written on a friend (friend), website (Twitter (registered trademark), blog, electronic bulletin board, SNS, etc.). The communication history database server 2 acquires a communication history from each terminal device 4 when a campaign is set in the user selection device 1.

  The call / mail history database server 3 is a server device including a database that stores a call history and a mail transmission / reception history for each user. The call history is a history of information related to the call of each terminal device 4, and the mail transmission / reception history is a history of information related to the mail of each terminal device 4. Information related to a call includes a call partner and a call time. In addition, the call in a personal computer is a call in an IP (Internet Protocol) telephone etc., for example. Further, the information regarding the mail includes the transmission source and reception time of the received mail, the transmission destination and transmission time of the transmitted mail, and the like. The mail in the personal computer is, for example, a web mail that is an electronic mail of a web application. When a campaign is set in the user selection device 1, the call / mail history database server 3 acquires a call history and a mail transmission / reception history from each terminal device 4.

  The user selection device 1 constituted by a computer selects a target person of a campaign to be executed based on information stored in the communication history database server 2 and information stored in the call / mail history database server 3.

  In this embodiment, the communication history database server 2 stores the communication history, and the call / mail history database server 3 stores the call history and mail transmission / reception history. However, the user selection device 1 stores the communication history. You may provide the database memorize | stored for every user and the database which memorize | stores the telephone call log | history and mail transmission / reception log | history for every user.

FIG. 2 is a block diagram illustrating a functional configuration of the user selection device 1 according to the present embodiment.
The user selection device 1 includes a campaign input unit 101, a candidate user selection unit 102, an acquaintance number calculation unit 103, a distribution generation unit 104, and a target user selection unit 105.
The campaign input unit 101 accepts campaign settings. Specifically, the campaign input unit 101 receives input of the contents of the campaign to be executed, and outputs the input contents of the campaign to the candidate user selection unit 102. The content of the campaign includes a keyword indicating the campaign.

  Candidate user selection section 102 selects a user having a hobby and a taste that matches the input campaign as a candidate user, and outputs a set of the selected candidate users to acquaintance calculation section 103. A candidate user is a node of a social network according to a campaign. Specifically, first, the candidate user selection unit 102 acquires the communication history of each user from the communication history database server 2. And the candidate user selection part 102 makes the user who browsed the website corresponding to a campaign a candidate user. A website that matches the campaign is a website that includes a keyword indicating the campaign. Further, the candidate user selection unit 102 sets this user as a candidate user when a keyword indicating a campaign is included in the content written by the user on Twitter (registered trademark), a blog, or the like.

  The acquaintance calculation unit 103 calculates the acquaintances of each candidate user in the set of candidate users and outputs the acquaintances of each candidate user to the distribution generation unit 104. Specifically, the acquaintance calculation unit 103 first acquires the call history and mail transmission / reception history of each candidate user from the call / mail history database server 3. Then, the acquaintance calculation unit 103 sets another candidate user B who has made a call or mail transmission / reception with the candidate user A as an acquaintance of the candidate user A. Further, the acquaintance calculation unit 103 acquires the communication history of each candidate user from the communication history database server 2. Then, the acquaintance calculation unit 103 sets another candidate user C who has communication relation with the candidate user A on the Internet as the acquaintance of the candidate user A. Users who have communication relations on the Internet are registered with friends (friends) on users who have exchanged messages such as chats over the Internet, and websites such as blogs and SNSs, for example. It is a user.

  The distribution generation unit 104 generates an order distribution of nodes in the campaign social network from the number of acquaintances of each candidate user, and outputs the generated order distribution to the target user selection unit 105. Here, the distribution generation unit 104 sets a node as a candidate user and sets the degree, which is the number of links of the node, as the number of acquaintances. The degree distribution p (k) is expressed by the following proportional expression (1) using the power exponent γ.

  However, k is the number of acquaintances.

  The target user selection unit 105 selects a user who executes the target campaign based on the degree distribution. Specifically, the target user selection unit 105 selects N candidate users (N is an integer equal to or greater than 1) with high importance U (k) as campaign target persons. The importance U (k) of each user is calculated by the following equation (2).

  However, f (k) and g (k) in the above-described equation (2) are expressed by the following equations (3) and (4).

Here, k _max is the maximum value of k.

  f (k) indicates the effect of information transmission according to the number of acquaintances k, is an index of the absolute value of the user, and is a value proportional to the number of acquaintances k. That is, f (k) is a function for increasing the importance of a user with a larger number of acquaintances.

  g (k) indicates an added value based on the scarcity of the user according to the number of acquaintances k, and is an index of the relative value of the user in the whole. In other words, g (k) is a function for increasing the importance of a user having a higher rarity.

Next, with reference to FIG. 3, the campaign object user selection process in the user selection apparatus 1 in this embodiment is demonstrated. FIG. 3 is a flowchart showing a procedure of campaign target user selection processing according to the present embodiment.
Hereinafter, a case where the campaign is “announcement of mobile application” will be described as an example.

First, the campaign input unit 101 receives an input of the content of the campaign “notification of mobile application” (step S101). The contents of the campaign “notification of mobile app” include the keyword “mobile app” indicating the campaign, keywords related to the mobile app to be notified (for example, the name and genre of the mobile app), and the content of the mobile app to be notified. Related keywords. These keywords can be set by an operator who sets a campaign. For example, when the mobile application is “Free Shogi (Board Game)”, keywords related to the contents of the mobile application include “Free Game”, “Board Game”, “Shogi”, “Free Application”, etc. Next, the candidate user selection unit 102 acquires the communication history of all users from the communication history database server 2 (step S102). The communication history database server 2 acquires a communication history from each terminal device 4 when the contents of the campaign are input in step S101.

Next, the candidate user selection unit 102 selects a candidate user based on the communication history acquired in step S102 (step S103). Specifically, the candidate user selection unit 102 sets a user that meets any of the following conditions 1 and 2 as a candidate user.
(Condition 1) The user has browsed a website that includes the keyword of the campaign “announcement of mobile app” (the keyword “mobile app” indicating the campaign, the keyword related to the mobile app, and the keyword related to the content of the mobile app) A certain user (Condition 2) A user who has a purchase history of “mobile app”

Next, the acquaintance calculation unit 103 acquires the call history and mail transmission / reception history of each candidate user from the call / mail history database server 3 (step S104). When a candidate user is selected in step S103, the call / mail history database server 3 acquires a call history and a mail transmission / reception history from the terminal device 4 of each candidate user.
Then, the number-of-people calculation unit 103 calculates the number of people in the set of candidate users for each candidate user (step S105). Specifically, the acquaintance calculation unit 103 counts the number of acquaintances with the candidate user that meets any of the following conditions A to C as acquaintances of the candidate user A.
(Condition A) Candidate user who has made a call with candidate user A (Condition B) Candidate user who has exchanged mail with candidate user A (Condition C) There was a communication relationship with candidate user A on the Internet Candidate user

Next, the distribution generation unit 104 generates the order distribution of each candidate user using the number of acquaintances as the order (step S106).
Finally, the target user selection unit 105 calculates the importance of each candidate user from the degree distribution generated in step S106, and selects a user who is the target of the campaign from the candidate users based on the calculated importance ( Step S107). Specifically, the target user selection unit 105 sets candidate users having high importance as target persons of the campaign.

  As described above, according to the present embodiment, a candidate network that matches a campaign is selected based on the communication history, and a social network is generated using the selected candidate user as a node. Network can be created. Thereby, a campaign target person who can effectively spread the campaign can be selected.

Further, a program for realizing each step shown in FIG. 3 is recorded on a computer-readable recording medium, and the program recorded on the recording medium is read into a computer system and executed, thereby selecting a user to be campaigned. Processing may be performed. Here, the “computer system” may include an OS and hardware such as peripheral devices.
Further, the “computer system” includes a homepage providing environment (or display environment) if a WWW system is used.
The “computer-readable recording medium” means a flexible disk, a magneto-optical disk, a ROM, a writable nonvolatile memory such as a flash memory, a portable medium such as a CD-ROM, a hard disk built in a computer system, etc. This is a storage device.

Further, the “computer-readable recording medium” means a volatile memory (for example, DRAM (Dynamic DRAM) in a computer system that becomes a server or a client when a program is transmitted through a network such as the Internet or a communication line such as a telephone line. Random Access Memory)), etc., which hold programs for a certain period of time.
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.

As described above, the embodiment of the present invention has been described in detail with reference to the drawings. However, the specific configuration is not limited to the above, and various design changes and the like can be made without departing from the scope of the present invention. It is possible to
For example, in this embodiment, the target person of the campaign is selected from the generated social network. However, in the case of a campaign such as point grant, the benefits to be provided such as changing the points to be given according to the importance are provided. You may use the produced | generated social network for the campaign etc. which are determined based on importance.
Further, in this embodiment, the target person of the campaign is selected based on the importance, but other selection methods based on the degree distribution are used, for example, a candidate user having a large degree (acquaintance) is targeted. May be.
In the present embodiment, the communication history database server 2 and the call / mail history database server 3 acquire each piece of information from each terminal device 4 after the campaign has been set. Alternatively, information regarding communication, information regarding a call, or information regarding a mail may be acquired from each terminal device 4 each time a mail is sent.
Further, the user selection device 1 may first generate a set of users to be a population and select candidate users from the generated population. For example, the attributes (for example, age, sex, occupation, etc.) of each user are stored in advance, and a set of users having attributes that match the campaign is set as a population. Specifically, if the campaign is “notification of mobile apps for students”, a set of users whose occupation is student is the population, and if the campaign is “notification of cosmetics”, the user is gender. For example, a set of

DESCRIPTION OF SYMBOLS 1 ... User selection apparatus 2 ... Communication history database server 3 ... Call / mail history database server 4 ... Terminal apparatus 101 ... Campaign input part 102 ... Candidate user selection part 103 ... Acquaintance calculation part 104 ... Distribution generation part 105 ... Target user selection Part

Claims (6)

A candidate user selection unit for selecting candidate users having hobbies and preferences that match the campaign from the users;
For each candidate user, a calculation unit that calculates the number of people in the set of candidate users;
A generation unit that generates a distribution of the number of acquaintances of each candidate user calculated by the calculation unit;
A target user selection unit that selects a target person of the campaign based on the distribution generated by the generation unit;
A user selection device comprising: A first storage unit for storing a browsing history of a website for each user;
The user selection apparatus according to claim 1, wherein the candidate user selection unit selects a user who browsed a website that matches the campaign as the candidate user. A second storage unit that stores a call history or a mail transmission / reception history for each user;
The user selection device according to claim 1, wherein the calculation unit sets another candidate user who has made a call or mail transmission / reception with the first candidate user as an acquaintance of the first candidate user. A third storage unit for storing a friend of each user on the website;
The calculation unit according to any one of claims 1 to 3, wherein the first candidate user and another candidate user who is a friend in the website are acquainted with the first candidate user. User selection device. A user selecting device selecting candidate users having hobbies and preferences that match the campaign from the users;
The user selection device calculating, for each candidate user, the number of acquaintances in the set of candidate users;
The user selection device generates a calculated distribution of the number of acquaintances of each candidate user;
The user selection device selecting a target person of the campaign based on the generated distribution;
A user selection method characterized by comprising: Selecting candidate users having hobbies and preferences that match the campaign from among the users;
Calculating the number of people in the set of candidate users for each candidate user;
Generating a distribution of the calculated number of acquaintances of each candidate user;
Selecting a target for the campaign based on the generated distribution;
A program that causes a computer to execute.

Images