JP4177018B2 - Content distribution method and program for causing computer to perform processing in the method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a content distribution method, and more particularly to a content distribution method for distributing content information such as video to be displayed on a display device (for example, a display on a table for eating and drinking) installed in a store.
[0002]
The present invention also relates to a program for causing a computer to perform processing in such a content distribution method.
[0003]
[Prior art]
In recent years, a large display may be installed in a store such as a restaurant to display a video or the like. On this display, movies, music videos, advertisements, etc. set in advance according to the time zone are displayed, and sound is output in accordance with the videos. A user who uses a store having such a display is unconsciously eating while watching the video displayed on the display.
[0004]
[Problems to be solved by the invention]
However, as described above, only the preset video is displayed on the display installed in a predetermined store, so that the video is displayed regardless of the user's intention, which is of interest to the user. You have to see no video or listen to audio. Therefore, there is a problem in that the store side cannot expect effects such as increasing the staying time in the store and increasing the order of meals by displaying images and advertisements on the display.
[0005]
Conventionally, a technology for recognizing a specific user's voice from a user's conversation and distributing content corresponding to the voice has been disclosed. In this case, a specific user's voice, that is, a specific user's voice is disclosed. Since only the user's conversation content affects the content distribution decision, there is a problem that the content distribution becomes uneven.
[0006]
Accordingly, an object of the present invention is to provide a content distribution method capable of distributing content related to a common topic in a conversation by a plurality of users, and a program for causing a computer to perform processing in the method. It is.
[0007]
[Means for Solving the Problems]
In order to solve the above problems, the present invention provides a content distribution method for distributing content information based on conversation information of a plurality of users by a computer, wherein the computer includes:
The first keyword extraction procedure for extracting a keyword from the conversation information and the keyword extracted by the first keyword extraction procedure are associated with the user and are the frequency of the keywords frequently appearing. Out Frequency information storage procedure for storing the degree information in the keyword DB, and among the above keywords, The total of keyword frequency information of each user is a keyword having a predetermined value or more, and the keyword is obtained from each of the plurality of users. A second keyword extraction procedure for extracting keywords generated at a more equal frequency from the keyword DB, and a content information DB in which content corresponding to the keyword is stored are searched and extracted by the second keyword extraction procedure. And a distribution procedure for distributing content information corresponding to the keyword.
[0008]
In such a content distribution method, keywords that are frequently used are managed based on conversation information of a plurality of users, and keywords that are generated at a similar frequency by a plurality of users are extracted and extracted. By distributing the content information based on keywords, it is possible to distribute content related to a common topic in a conversation by a plurality of users.
[0009]
In order to solve the above problems, the present invention provides a program for causing a computer to perform the processing according to the content distribution method. And You can also
[0010]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, embodiments of the present invention will be described with reference to the drawings.
[0011]
FIG. 1 shows a system configuration diagram in a content distribution method according to an embodiment of the present invention. In FIG. 1, a content distribution system 10 in the content distribution method is connected to a plurality of stores 2 to 2N that perform content distribution processing, a store server 20 provided in the store 2, and the store server 20 via a LAN 15. This is a system including a plurality of user terminals 30 operated by a user who uses the content server 40, a content server 40 that distributes content, and the like.
[0012]
In FIG. 1, a store server 20 and a content server 40 in the store 2 are server computers, and each component is controlled by a CPU (central processing unit). The CPU performs processing in the content distribution method according to the program stored in the memory unit. The network server 1 is connected to the store server 20 of the store 2, the user terminal 30, and the content server 40. In addition, the user side terminal 30 may be connected only to the store server 20 and the content information stored in the store server 20 may be distributed.
[0013]
The store server 20 mainly includes an installer 11 that installs a program for executing processing related to content distribution processing, a communication control processing unit 12 that controls data communication, a display control processing unit 13 that controls display of data, An input / output control processing unit 14 that controls input / output of data, a voice information recognition processing unit 21 that recognizes a specific user based on voice information acquired by the user terminal 30, and a user based on the voice information The voice information determination processing unit 22 that determines a specific user of the side terminal 30, the keyword extraction processing unit 23 that extracts a keyword for distributing content based on the voice information, and the keyword extraction processing unit 23 A content distribution instruction processing unit 24 that issues a content distribution instruction based on a keyword, and a key that stores a keyword from audio information. Having a word DB 25, a personal identification information DB26 to the personal identification information is stored for identifying the individual user utilizing user-side terminal 30, and a threshold table 27 for the extraction of keywords.
[0014]
The communication control processing unit 12 is a processing unit that controls transmission / reception data via the network 1 and includes a communication unit for connecting to a network. The display control processing unit 13 includes a display unit such as a display, and controls display data to display on the display unit. The input / output control processing unit 14 controls input / output of data by controlling an input unit such as a mouse and a keyboard and an output unit such as a printer. The installer 11 reads the program from the CD-ROM 50 on which the program for realizing the content distribution method according to the embodiment of the present invention is recorded, and installs the read program in an auxiliary storage device or the like, for example. To do. Then, when processing in the content distribution method described later is started, the CPU starts the processing according to the program installed in the auxiliary storage device. The medium for storing the program is not limited to a CD-ROM, and any medium that can be read by a computer may be used.
[0015]
The voice information recognition processing unit 21 recognizes a specific user based on the voice information acquired by the user terminal 30 and the personal identification information stored in the personal identification information DB 26, and the personal information as a result of the recognition is obtained. Processing such as updating is performed. The voice information determination processing unit 22 determines a specific user of the user side terminal 30 based on the voice information and the personal information stored in the personal identification information DB 26, and based on the determination result, the keyword from the voice information. Information is stored in the keyword DB 25. The keyword extraction processing unit 23 extracts keywords for distributing content based on the voice information, the keyword information stored in the keyword DB 25, and the threshold value table 27 for extracting keywords. The content distribution instruction processing unit 24 issues a content distribution instruction to the content server 40 based on the keywords extracted by the keyword extraction processing unit 23. The keyword information stored in the keyword DB 25 is shown in FIG. 11, FIG. 12 and FIG. 14, the personal identification information stored in the personal identification information DB 26 is shown in FIG. 10, and the threshold table 27 is shown in detail in FIG. explain.
[0016]
The plurality of user terminals 30 provided in the store 2 include a display control processing unit 31 and a voice information acquisition processing unit 32. The display control processing unit 31 includes a display unit such as a display, and performs processing of controlling content information transmitted from the content server 40 or the store server 20 and displaying the content information on the display unit. The voice information acquisition processing unit 32 includes a voice acquisition unit such as a microphone, and performs a process of acquiring voice information when the user is talking and transmitting the voice information to the store server 20. This audio information will be described in detail with reference to FIG.
[0017]
The content server 40 includes a content DB 41 that stores a variety of content information and a distribution processing unit 42. The distribution processing unit 42 receives the content distribution instruction from the store server 20 provided in each store 2, and based on the content distribution instruction and the content information stored in the content DB 41, the predetermined content information is stored on the user side terminal. 30 or to the store server 20.
[0018]
Next, the processing procedure of the content distribution method will be described. FIG. 2 is a flowchart for explaining the content distribution processing. First, in the process of step S10, whether or not the personal identification information has been acquired by the store server 20, that is, the personal identification information generated from the voice information in the conversation for a certain period at the user side terminal 30 is the personal identification. It is determined whether it is stored in the information DB 26. If the personal identification information has not been acquired in the process of step S10, the voice information recognition process is performed by the voice information recognition processing unit 21 of the store server 20 in the process of step S11. The voice information recognition processing procedure will be described in detail with reference to FIG. If the personal identification information has already been acquired in the process of step S10, the voice information determination processing unit 22 and the keyword extraction processing unit 23 of the store server 20 use the voice information, the personal identification information, and the threshold value in the process of step S12. Based on the values in the table 27, voice information determination and keyword extraction processing are performed. The voice information determination processing procedure will be described in detail with reference to FIGS. 4 and 5, and the frequent keyword extraction processing procedure will be described in detail with reference to FIG. In the process of step S13, the keyword extracted by the content distribution instruction processing unit 24 of the store server 20 is transmitted to the content server 40 to instruct content distribution. The content distribution instruction processing procedure will be described in detail with reference to FIG. In the process of step S14, the store server 20 determines whether the content distribution process is finished. If the content distribution process has not ended in the process of step S14, the process from step S12 is repeated.
[0019]
In step S14, whether or not the content distribution process is completed is determined based on a period preset in the store server 20. Further, the operator of the store server 20 may determine the end of the content distribution process.
[0020]
FIG. 3 is a flowchart for explaining the voice information recognition process. In FIG. 3, the voice information acquired by the voice information acquisition processing unit 32 of the user side terminal 30 is received by the voice information recognition processing unit 21 of the store server 20 in the process of step S20.
[0021]
FIG. 9 shows a voice information table. In FIG. 9, the audio information table 320 acquired by the audio information acquisition processing unit 32 stores the audio information when the user has a conversation at the table where the user terminal 30 is installed. Is transmitted to the store server 20.
[0022]
In step S21, phoneme information and utterance level information are analyzed from the voice information received by the voice information recognition processing unit 21. In step S22, the voice information recognition processing unit 21 acquires the personal identification information stored in the personal identification information DB 26 and performs personal identification processing. In step S23, the voice information recognition processing unit 21 determines whether the personal identification information includes data approximate to the voice information analyzed in step S21. When there is no data approximate to the voice information analyzed in the personal identification information in the process of step S23, the voice information analyzed by the voice information recognition processing unit 21 in the process of step S24 is stored in the personal identification information DB 26 as the personal identification information.
[0023]
FIG. 10 is a diagram showing a table stored in the personal identification information DB. In FIG. 10, the table 260 stored in the personal identification information DB 26 includes personal identification information that identifies the user of the user side terminal 30, phonological information that is analyzed based on voice information, and utterance level information. Stored. For example, the table 260 stores phonological information “data A” and utterance level information “level 1” in association with the personal identification information “0001”.
[0024]
If there is data similar to the voice information analyzed in the personal identification information in the process of step S23, the voice information of all who are talking on the user terminal 30 by the voice information recognition processing unit 21 in the process of step S25, That is, it is determined whether or not personal identification information has been acquired. If the personal identification information of all members is acquired in the process of step S25, the voice information recognition process is terminated. If all the personal identification information has not been acquired in the process of step S25, the process from step S20 is repeated.
[0025]
In the process of step S25, whether or not all the personal identification information has been acquired is determined based on whether or not all the personal identification information has been acquired when there is no new registration in the store server 20 in a preset period. I do. Further, it may be determined whether or not the operator of the store server 20 has acquired all the personal identification information.
[0026]
4 and 5 are flowcharts for explaining the audio information determination process. In FIG. 4, first, in the process of step S <b> 30, the voice information determination processing unit 22 receives voice information from the user side terminal 30. In step S31, phoneme information and utterance level information are analyzed from the voice information received by the voice information determination processing unit 22. In the process of step S32, based on the personal identification information stored in the personal identification information DB 26 by the voice information determination processing unit 22, a personal determination process is performed to determine which user has issued the voice information. In the process of step S33, the voice information determination processing unit 22 extracts keywords from the voice information. The frequency information and standard deviation information of the keyword information stored in the keyword DB 25 are stored based on the keyword extracted by the voice information determination processing unit 22 in the process of step S34 and the individual determination result by the process of step S32. To do. The frequency information is a frequency of keywords that occur in the voice information of each user. Also, the standard deviation information is the keyword generated in the voice information of each user for each keyword. Frequent Standard deviation value indicating variance. That is, a keyword having a small standard deviation value is determined as a keyword that has occurred with a similar frequency to a plurality of users.
[0027]
For example, in the process of step S33, the voice information determination processing unit 22 extracts the keywords “Okinawa”, “Sea”, and the like from the voice information “Okinawa sea was beautiful” in the voice information table 320 shown in FIG. After extracting the keywords “Okinawa”, “Ocean”, etc., 1 is added to each frequency information of the keywords “Okinawa” and “Ocean” stored in the keyword DB 25 and updated by the processing of Step S34. In the process of step S35, it is determined whether or not a series of voice information acquisition is completed by the voice information determination processing unit 22, that is, whether or not a series of voice information acquisition issued by the user of the personal determination result is completed. If the series of voice information acquisition is not completed in the process of step S35, the process from step S33 is repeated. When a series of voice information acquisition is completed in the process of step S35, it is determined whether or not a predetermined time has elapsed since the voice information determination processing unit 22 starts acquiring the voice information in the process of step S36. If the predetermined time has not elapsed since the start of the acquisition of the voice information in the process of step S36, the process from step S30 is repeated. When a predetermined time has elapsed since the start of the acquisition of audio information in the process of step S36, the audio information determination processing unit 22 determines whether or not the content distribution process is the first time in the process of step S37.
[0028]
If the content distribution process is the first time in the process of step S37, the keyword extraction processing unit 23 performs a process of extracting a keyword with high frequency information based on the keyword information in the keyword DB 25 in the process of step S40 shown in FIG. The frequent keyword extraction processing procedure in step S40 will be described in detail with reference to FIG. In the process of step S41, the extracted frequent keyword information is stored in the keyword DB 25 by the voice information determination processing unit 22. In the process of step S42, the voice information determination processing unit 22 stores the frequency information and standard deviation information of the frequent keywords in the keyword DB 25. In the processing of step S43, the voice information determination processing unit 22 sets the frequent keyword distribution flag to ON and stores it in the keyword DB 25.
[0029]
If the content distribution process is not the first time in the process of step S37 shown in FIG. 4, the voice information determination processing unit 22 reads the frequency information of the extracted keywords from the keyword DB 25 in the process of step S44 shown in FIG. In step S45, the voice information determination processing unit 22 extracts a common keyword from the keyword information stored in the keyword DB 25, that is, the keyword extracted the same time as the keyword extracted this time. In the process of step S46, it is determined whether or not the difference between the frequency information of the keyword extracted this time and the previously extracted keyword by the voice information determination processing unit 22 is smaller than the threshold value set in the threshold value table 27. If the difference in the frequency of occurrence of each keyword is smaller than the threshold value in step S46, the processing from step S40 is performed. If the difference in the frequency information of each keyword is larger than the threshold value in step S46, the voice information determination processing unit 22 sets the frequent keyword distribution flag to OFF and stores it in the keyword DB 25 in step S47.
[0030]
FIG. 6 is a flowchart for explaining frequent keyword extraction processing. In FIG. 6, first, in the process of step S <b> 50, one word is read from the keyword information stored in the keyword DB 25 by the keyword extraction processing unit 23. In step S 51, the keyword extraction processing unit 23 determines whether the keyword standard deviation information is equal to or greater than a threshold value set in the threshold value table 27. If the standard deviation information of the keyword is equal to or smaller than the threshold value in step S51, the keyword extraction processing unit 23 determines whether the frequency information is equal to or larger than the threshold value in step S52. In the process of step S52, the frequency information is below a threshold value. under If so, the keyword information read by the keyword extraction processing unit 23 is temporarily stored in the keyword DB 25 in the process of step S53.
[0031]
In the processing of step S54, it is determined whether or not all keyword information stored in the keyword DB 25 has been searched by the keyword extraction processing unit 23. When all the keyword information stored in the keyword DB 25 is searched in the process of step S54, the keyword extraction processing unit 23 extracts the keyword having the lowest standard deviation information in the process of step S55 and temporarily stores it in the keyword DB 25. .
[0032]
FIG. 7 is a flowchart for explaining the content distribution instruction process. In FIG. 7, first, in the process of step S <b> 60, the distribution flag stored in the keyword DB 25 is read by the content distribution instruction processing unit 24. In step S61, it is determined whether the distribution flag read by the content distribution instruction processing unit 24 is ON or OFF. If the distribution flag is OFF in the process of step S61, the process from step S60 is repeated. If the distribution flag is ON in the process of step S61, the content distribution instruction processing unit 24 reads the keyword information stored in the keyword DB 25 in the process of step S62. In step S63, the content distribution instruction processing unit 24 transmits a content distribution instruction to the content server 40.
[0033]
FIG. 8 is a flowchart for explaining the distribution process. In FIG. 8, first, the distribution processing unit 42 of the content server 40 receives a content distribution instruction from the store server 20 in the process of step S <b> 70. In step S71, the distribution processing unit 42 searches for content information corresponding to the keyword based on the content information corresponding to the plurality of keywords stored in the content DB 41 and the keyword of the received content distribution instruction. In the process of step S72, the distribution processing unit 42 determines content information to be distributed. In step S73, the distribution processing unit 42 distributes the determined content information to the user terminal 30.
[0034]
Next, the keyword information and threshold value table stored in the keyword DB 25 will be described.
[0035]
FIG. 11 is a diagram showing a table stored in the keyword DB. In FIG. 11, a table 250 stored in the keyword DB 25 stores extracted keyword information, standard deviation information of the keyword, personal identification information that issued the keyword, and frequency information of the keyword. . For example, the table 250 stores keyword information “Okinawa”, standard deviation information “40”, personal identification information “0001”, and frequency information “85”. The table 250 stores the total frequency information of the extracted keywords.
[0036]
FIG. 12 is a diagram illustrating a table stored in the keyword DB. 12, the table 251 stored in the keyword DB 25 includes keyword information whose standard deviation information is smaller than the threshold and whose frequency information is larger than the threshold based on the keyword information shown in FIG. 11 by the keyword extraction processing unit 23. It is a stored table. For example, the frequency information “50/175”, the standard deviation information “3”, the keyword information “Awamori”, and the distribution plug “OFF” are stored in the table 251.
[0037]
FIG. 13 shows a threshold table. In FIG. 13, the threshold value table 27 stores preset threshold values. For example, the threshold value table 27 stores the threshold value “0.25” of the frequency information, the threshold value “2” of the standard deviation information, and the threshold value “0.1” of the difference between the previous frequency information and the current frequency information. .
[0038]
FIG. 14 is a diagram illustrating a table stored in the keyword DB. In FIG. 14, in the table 252 stored in the keyword DB 25, the keyword information of the distribution flag ON is stored by the content distribution instruction processing unit 24 from the extracted keyword information with high frequency information shown in FIG. For example, the table 252 stores the frequency information “25/175”, the standard deviation information “1.5”, and the distribution flag “ON”.
[0039]
In the above content distribution method, it is not limited to voice information, but is generated by a plurality of users at the same frequency based on character information generated by a plurality of users using the chat function on the Internet. You may make it extract a common keyword.
[0040]
In the content distribution method, advertisement information based on the extracted keyword may be distributed.
[0041]
As described above, in the content distribution method, based on the voice information of conversations among a plurality of users, the generated keywords with high frequency information, the standard deviation information for each keyword, the frequency information for each user, An identifier indicating a content information distribution instruction based on the extracted keyword is extracted from a plurality of managed keywords that are generated at a similar frequency by a plurality of users based on a threshold value. By storing in the keyword, it is possible to distribute content related to a common topic in a conversation by a plurality of users.
[0042]
In addition, in the above content distribution method, when the difference between the frequency information of the keyword that is instructed to distribute the previous content and the frequency information of the keyword extracted this time is larger than the threshold, the distribution of the content information that is different from the previous time By instructing, it is possible to distribute content according to the conversation situation of the user.
(Supplementary Note 1) A content distribution method for distributing content information based on conversation information of a plurality of users,
A management procedure for managing frequently used keywords based on the conversation information,
A keyword extraction procedure for extracting keywords generated by the plurality of users at the same frequency among the keywords by the management procedure;
A distribution procedure for distributing the content information based on the keyword extracted by the keyword extraction procedure.
(Additional remark 2) The said management procedure manages the standard deviation information which shows the standard deviation of the said keyword in association with every said keyword,
The content distribution method according to appendix 1, wherein the keyword extraction procedure extracts keywords whose standard deviation information is equal to or less than a threshold value.
(Additional remark 3) The said management procedure manages the frequency information which shows the frequency which generate | occur | produced for every said user in association with every said keyword,
The content distribution method according to appendix 1 or 2, wherein the keyword extraction procedure extracts keywords whose frequency information is equal to or greater than a threshold value.
(Additional remark 4) The said keyword extraction procedure has a judgment procedure which judges whether the difference of the frequency information of the keyword extracted last time and the frequency information of the keyword extracted this time is below a threshold value,
The content delivery method according to appendix 3, wherein the delivery procedure includes a delivery disabling setting procedure for setting information indicating an instruction not to deliver the content information based on a determination result obtained by the determination procedure.
(Additional remark 5) Furthermore, the table which shows the threshold value of the said frequency information, the threshold value of the said standard deviation information, and the threshold value of the difference between the frequency information of the keyword extracted last time and the frequency information of the keyword extracted this time The content distribution method according to any one of appendices 2 to 4, further comprising a threshold management procedure for management.
(Supplementary note 6) The content distribution method according to any one of supplementary notes 1 to 5, wherein the conversation information is voice information generated orally by the plurality of users.
(Supplementary note 7) The content distribution method according to any one of supplementary notes 1 to 6, wherein the conversation information is character information input online by the plurality of users on the Internet.
(Supplementary note 8) The content distribution method according to any one of supplementary notes 1 to 7, wherein the content information is advertisement information based on the keyword.
(Supplementary Note 9) A program for causing a computer to perform processing in a content distribution method for distributing content information based on conversation information of a plurality of users,
A management procedure for managing frequently used keywords based on the conversation information,
A keyword extraction procedure for extracting keywords generated by the plurality of users at the same frequency among the keywords by the management procedure;
A distribution procedure for distributing the content information based on the keyword extracted by the keyword extraction procedure.
(Supplementary Note 10) A storage medium storing a program for causing a computer to perform processing in a content distribution method for distributing content information based on conversation information of a plurality of users,
A management procedure for managing frequently used keywords based on the conversation information,
A keyword extraction procedure for extracting keywords generated by the plurality of users at the same frequency among the keywords by the management procedure;
And a distribution procedure for distributing the content information based on the keyword extracted by the keyword extraction procedure.
[0043]
【The invention's effect】
As described above, according to the present invention, frequently used keywords are managed based on conversation information of a plurality of users, and keywords generated by a plurality of users at the same frequency are extracted and extracted from the keywords. By distributing the content information based on the keywords, it is possible to distribute content related to a common topic in a conversation by a plurality of users.
[Brief description of the drawings]
FIG. 1 is a system configuration diagram in a content distribution method according to an embodiment of the present invention.
FIG. 2 is a flowchart for explaining the entire content distribution process;
FIG. 3 is a flowchart for explaining voice information recognition processing;
FIG. 4 is a flowchart for explaining audio information determination processing;
FIG. 5 is a flowchart for explaining audio information determination processing;
FIG. 6 is a flowchart for explaining frequent keyword extraction processing;
FIG. 7 is a flowchart for explaining content distribution instruction processing;
FIG. 8 is a flowchart for explaining distribution processing;
FIG. 9 is a diagram showing a table of audio information.
FIG. 10 is a diagram showing a table stored in a personal identification information DB.
FIG. 11 is a diagram showing a table stored in a keyword DB.
FIG. 12 is a diagram showing a table stored in a keyword DB.
FIG. 13 is a diagram showing a threshold table.
FIG. 14 is a diagram showing a table stored in a keyword DB.
[Explanation of symbols]
1 network
2-2N store
10 Content distribution system
11 Installer
12 Communication control processing unit
13 Display control processor
14 Input / output control processor
15 LAN
20 store server
21 Voice information recognition processor
22 Voice information determination processing section
23 Keyword extraction processing unit
24 Content distribution instruction processing unit
25 Keyword DB
26 Personal identification information DB
27 Threshold table
30 User terminal
31 Display control processing unit
32 Voice information acquisition processing unit
40 content server
41 Content DB
42 Distribution processing section
50 CD-ROM

Claims (2)

A content distribution method for distributing content information based on conversation information of a plurality of users by a computer,
The computer is
A first keyword extraction procedure for extracting keywords from the conversation information;
The relative first keyword extracting keywords extracted in step, in association with the user, and the frequency information storage procedure for storing the frequent output of information is a keyword frequent degree keyword DB,
Among the above keywords, keywords whose total keyword frequency information of each user is equal to or greater than a predetermined value, and keywords that are generated from each of the plurality of users at a more equal frequency are obtained from the keyword DB. A second keyword extraction procedure to extract;
A content information DB storing content corresponding to a keyword, and executing a distribution procedure for distributing content information corresponding to the keyword extracted by the second keyword extraction procedure Delivery method. A program for causing a computer to perform processing in a content distribution method for distributing content information based on conversation information of a plurality of users,
A first keyword extraction procedure for extracting keywords from the conversation information;
The relative first keyword extracting keywords extracted in step, in association with the user, and the frequency information storage procedure for storing the frequent output of information is a keyword frequent degree keyword DB,
Among the above keywords, keywords whose total keyword frequency information of each user is equal to or greater than a predetermined value, and keywords that are generated from each of the plurality of users at a more equal frequency are obtained from the keyword DB. A second keyword extraction procedure to extract;
Searching a content information DB in which content corresponding to a keyword is stored, and causing a computer to perform a distribution procedure for distributing content information corresponding to the keyword extracted by the second keyword extraction procedure. Program to do.

Images