JP2014222474A - Information processor, method and program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide information processing of information selection such that importance to other users is also reflected by making information to be selected correctly reflect user's interest.SOLUTION: An information processor includes presentation candidate information acquisition means 2 for acquiring presentation candidate information tied to any of a plurality of genres, presentation candidate information evaluation means 3 for evaluating importance in each acquired presentation candidate information, and presentation information selection means 5 for correcting the importance in each presentation candidate information on the basis of an estimation value of the degree of user's interest tied to each genre, calculating a probability of selecting presentation candidate information as presentation information on the basis of the corrected importance, and selecting presentation information on the basis of the probability.

Description

  The present invention relates to an information processing apparatus, method, and program, and more particularly, to information processing that acquires a lot of information and selects some of them.

  Patent Document 1 discloses a system in paragraph 0054 that automatically acquires a user's preference while using a portable information terminal and always provides communication with other users based on the latest preference. It is disclosed. For learning about user preferences, for example, in paragraph 0025, static information such as the user's age and gender, and dynamic information such as when and where the user visits and what spots and information are interested To generate data consisting of:

JP 2001-265809 A

  The amount of information distributed using RSS on the World Wide Web is enormous, and it is usually impossible to present all of the acquired information to the user at one time. Therefore, some of the acquired information is selected and presented to the user. Also, when a search is performed by passing a keyword to a search engine and a very large number of search results are hit, the search results are usually selected. For example, the top dozens are presented to the user in descending order of relevance determined based on some algorithm.

  As such information selection or selection, in addition to a method based on a high degree of relevance, for example, a method using a user's preference is conventionally performed. The user is allowed to select in advance what kind of genre he / she is interested in, and only the information of the genre is displayed. Information on the selected genre is displayed, and information on a genre that has not been selected is not displayed.

  However, in this case, there are the following problems. It is difficult to select the genre you are interested in beforehand because the genre you are interested in may vary depending on the individual, but the user who sets the preference may not be aware of his / her preference accurately. There is a possibility. Also, in order to increase the degree of matching between preference settings and user interests, it is necessary to divide and manage many genres. However, such a method of asking the user about the degree of interest in many genres and selecting them is a very complicated operation and is difficult to use. In addition, it is difficult for the system to follow the change in the degree of interest of the user for each genre well.

  In addition, since information on a genre that has not been selected is not presented regardless of the importance level or general recognition level of information, there is a problem that the knowledge and information obtained by the user are so-called “takotsubo”. Then, there is a problem that there is a possibility of getting bored with the application for selecting information.

  In addition, based on the history of content browsed by the user, if it is determined that a genre browsed a predetermined number of times is interested and a genre that has not reached the genre is not interested, the following problems arise.

  In the initial stage of estimating the user's interest, it is possible to indicate that the genre is not so much of an interest, but it happens to show the user's interest, or that the genre is of interest to the user, but the user doesn't show an interest. Possible as sex. On the other hand, if the display of the content of the genre is stopped because it has not shown interest, the opportunity for estimating the interest in the genre is lost. If the threshold value is set large to alleviate this problem, reflection of the user's interest will be delayed.

  The present invention has been made in view of the above-described problems, and information selection information in which selected information correctly reflects the interest of the user and also reflects the importance for other users. The purpose is to provide processing.

  One configuration for achieving the above object is a presentation candidate information obtaining unit that obtains presentation candidate information associated with any of a plurality of genres, and presentation candidate information that evaluates the importance of each obtained presentation candidate information. The evaluation means and the importance for each candidate presentation information are corrected based on the estimated value of the degree of interest of the user associated with each genre, and the candidate presentation information is selected as the presentation information based on the corrected importance. And a presentation information selection means for selecting presentation information based on the probability.

  ADVANTAGE OF THE INVENTION According to this invention, it becomes possible to provide the information processing of information selection that the selected information correctly reflects a user's interest and the importance for other users.

It is a block diagram which shows the function structure of the 1st Embodiment of this invention. It is a figure for demonstrating the relationship of each information in the said 1st Embodiment. It is a flowchart figure which shows the procedure of the information processing in the said 1st Embodiment. It is a figure which shows the external appearance structural example of the 2nd Embodiment of this invention. It is a figure which shows the hardware & software structural example of the said 2nd Embodiment. It is a block diagram which shows the function structure of the said 2nd Embodiment. It is a figure which shows an example of the information source table 123 of FIG. It is a figure which shows an example of the interest table 122 of FIG. It is a figure which shows an example of the screen which the information display means of FIG. It is a figure which shows another example of the interest table 122 of FIG. It is a figure which shows an example of the screen displayed on the display apparatus 14 of FIG. It is a figure which shows an example of the screen displayed on the display apparatus 14 of FIG. It is a figure which shows the example of the change degree of the interest degree index | exponent by user operation in the said 2nd Embodiment.

<First Embodiment>
FIG. 1 is a block diagram showing a functional configuration of the first embodiment of the present invention. As illustrated, the information processing apparatus 1 according to this embodiment includes a presentation candidate information acquisition unit 2, a presentation candidate information evaluation unit 3, a genre interest coefficient storage unit 4, a presentation information selection unit 5, and a presentation information evaluation. Means 6. The information processing apparatus 1 inputs presentation candidates acquired from an external apparatus and information related thereto, performs presentation information selection processing together with an interest coefficient stored therein, and outputs the information as presentation information. When feedback to this output is obtained, it is reflected in the interest coefficient.

  FIG. 2 is a diagram for explaining the relationship between each piece of information handled in the present embodiment. In the present embodiment, a lot of information is acquired as “presentation candidates” from the outside, and some of them are selected or selected as “presentation information”. Typical examples of presentation candidates and presentation information include news articles distributed in a format such as RSS or Atom. However, the format need not be limited to these, and the information content need not be a news article.

  In the present embodiment, all “presentation candidates” belong to some “genre”. Here, considering that a typical example of “presentation candidate” is a news article, it is referred to as “genre”. However, it may be referred to as “category” or the like, or “presentation candidate” from another viewpoint that is not a genre. It may be classified when looking at. As an example of such classification, for example, “the length of the presentation information” can be considered. In this case, information processing such as sorting of articles acquired by the length of the article is performed by dividing into “long article” and “short article”.

  The “genre” in the present embodiment is a genre to which a news article belongs, and examples of such a genre include international, domestic, political, economic, entertainment, sports, and columns. The “genre” may be hierarchical.

In the present embodiment, an interest coefficient is set and stored for every “genre” for each user. The interest coefficient represents how much the user is interested in the genre, and is used for selecting articles by correcting general importance evaluated for each article. The interest coefficient is expressed as “A _j ”. On the other hand, the presentation candidate is represented as “S _i ”. A (S _i ) represents the interest coefficient of the genre to which the presentation candidate i belongs. The interest coefficient A _j has an initial value of 1.

Refer to FIG. 1 again. FIG. 3 shows a flow of information selection processing according to the present embodiment. 1 and 3, the presentation candidate information acquisition unit 2 acquires a presentation candidate S _i from the outside and inputs it to the presentation information selection unit 5 (S101). On the other hand, the presentation candidate information evaluation unit 3 evaluates the obtained presentation candidate S _i and inputs the importance f (S _i ) of the presentation candidate S _i to the presentation information selection unit 5 (S102).

For example, when the information processing apparatus 1 acquires the RSS of the external device as the presentation candidate S _i , the presentation candidate information evaluation unit 3 calculates the average number of accesses per certain time by other users for each article included in the RSS. Assess the importance of the article. Information about the presentation information such as the average number of accesses is acquired from an external device by some method. For example, it is conceivable that the news distribution site is set to automatically acquire an access ranking that is updated every few minutes. Of course, the information for evaluating the importance is not limited to the average number of accesses. For example, the number of citations in a social networking service (hereinafter referred to as SNS) per certain time (the number of retweets in Twitter (registered trademark, a service provided by US Twitter)) may be used.

  The genre interest coefficient storage means 4 stores an interest coefficient indicating the degree of interest of the user for each genre. On the other hand, the presentation information evaluation unit 6 responds to the reaction performed by the user with respect to the presentation information selected by the presentation information selection unit 5, and the degree of interest of the user with respect to the genre to which the selected presentation information belongs, that is, , Change the coefficient of interest.

  In other words, it can be said that the above-described function of the presentation information evaluation unit 6 estimates the interest in the genre to which the presentation information belongs by the user's action. When the presentation information shows only a part of the news article, whether the entire information is displayed, whether the presentation information can be looked back later, and quote the presentation information Then, the presentation information evaluation means 6 determines the amount by which the degree of interest is increased / decreased for each action such as whether it is shared with other users or whether it is deleted without being displayed.

  Further, the amount to be increased or decreased may be changed according to the target information. For example, if the entire information that is estimated to be low in importance and the degree of interest of the user is displayed, the degree of interest is greatly increased and the difference between the two is filled in compared to the case where the information is estimated to be high. Should. Similarly, if the presentation information that was estimated to have a high importance level and a high degree of user interest is judged to be unnecessary and is thrown into the trash, it is more interesting than if it was estimated to be low. The degree should be greatly reduced.

The presentation information selection unit 5 corrects the importance f (S _i ) of the input presentation candidate S _i by the interest coefficient A _j input from the genre interest coefficient storage unit 4 (S103). In the present embodiment, the specific method of correction is to calculate the estimated interest level A (S _i ) × f (S _i ).

Moreover, the presentation information selecting means 5, in accordance with the presented candidate S _i of the estimated interest degree _{A (S i) × f (} S i), presenting candidate S _i to calculate the probability P _i that is selected (S104). The specific method of calculation is according to the following formula in this embodiment. Although correction is performed by integration in the following equation, it is naturally not limited thereto. Addition may be used. However, in the following, in order to obtain the probability, it is desirable that the number is a positive number. Here, an integration that can satisfy the condition is used.

The presentation information selection means 5 presents probabilistic information based on the probability P _i calculated in this way (S105). Here, the presentation of probabilistic information is performed by selecting and displaying the number of pieces of information requested for presentation according to the presentation probability P _i for each presentation candidate information. Do not select the same information repeatedly. For example, presentation information is selected one by one, and the selected presentation candidate is removed each time selection is performed, and the probability is recalculated.

  By presenting probabilistic information as described above, even if it is a presentation candidate belonging to a genre with a low degree of interest for the user according to the user's past history, the possibility of being selected is low but zero Never become. Therefore, it is determined that it should not be presented to the user with a slight difference, and thereafter an erroneous determination that the user's interest level is not measured at all for the genre is not performed. In addition, not only information that is estimated to have a high degree of interest of the user but also a presentation candidate that many people are interested in is likely to be selected. Therefore, even if the information is unexpected for the user, what is important for the other users is reflected in the selection.

In the present embodiment, it is preferable that the range of change of the interest coefficient A _j by the presentation information evaluation unit 6 is 0 <A _j . According to this configuration, since the presentation probability P _i does not become zero, there is a possibility that even a genre that is not interesting to the user at the time when the selection process is performed will be selected in the future. Therefore, it is possible to follow changes in user interest in the future. Conversely, for a genre that is prohibited from being presented by the user, the presentation probability can be reduced to zero by setting A _i to zero.

<Second Embodiment>
Hereinafter, the second embodiment will be described with reference to FIGS. As described in the first embodiment, this embodiment is an information acquisition application including information processing that acquires RSS or feed including news articles from the Internet cloud and selects and presents some of them. It is an embodiment to which various technical ideas are applied.

  In FIG. 4, the example of an external appearance structure of this embodiment is shown. An information processing apparatus 100 according to the present embodiment is a personal device such as a personal computer (hereinafter, mainly referred to as “PC”), a slate type PC, a tablet type PC, a smartphone, or a portable information terminal (Personal Digital Assistance: PDA). Yes, communication with the server 200 on the cloud is possible.

  As the server 200 on the cloud, for example, SNS servers 201 and 202 that provide social networking services (hereinafter referred to as SNS), news distribution that provides a service that distributes news in a format such as RSS or Atom There are servers 203 and 204 and servers that provide other services. Various other services are conceivable. For example, a weather information distribution server 205 that distributes weather information, a short text posting server 206 that provides a short text posting site that can post relatively short texts and photos, etc., electronic commerce There are EC servers 207 and 208 that provide services.

  FIG. 5 shows a hardware & software configuration example of the information processing apparatus 100. As illustrated, the information processing apparatus 100 includes a central processing unit 10 as arithmetic control means, and includes a primary storage 11 and a secondary storage 12 as storage means. As input / output means, a network interface 13 for communicating with a host on the network, a display device 14 for performing display output, and an operation input unit 15 for receiving an operation input by a user are provided.

  The secondary storage 12 stores an information acquisition application 121, an interest table 122, and an information source table 123. The information acquisition application 121 is an application software program, and is provided from a server (for example, the EC server 207) on the network via a computer-readable optical or magnetic storage medium or via a network.

  The information acquisition application 121 is a composite application that provides several functions including the functions described below when executed by the central processing unit 11. In providing the following functions, an operating system or an external application library may be used.

<News Reader>
The information acquisition application 121 functions as a news reader by having the following configuration. FIG. 6 shows a functional block diagram. The information processing apparatus 100 includes the functional blocks shown in FIG. 6 through the cooperation of hardware and software programs as shown in FIG. That is, the information processing apparatus 100 includes an information acquisition unit 301, an information selection unit 303, an all information acquisition unit 304, an information display unit 305, an interest estimation unit 306, and an interest table display editing unit 307.

  The information source table 123 is a table in which information about an information source from which the information acquisition application 121 acquires information is stored. Examples of information sources include SNS, news distribution service, weather information distribution service, and electronic commerce service. FIG. 7 shows an example of the information source table 123.

  As shown in FIG. 7, each service is associated with information for accessing the service. A typical example of information for accessing a service is a URI (Uniform Resource Indicator) of a service site on the Internet, but is not limited thereto. Each service is associated with a service type.

  Each service may be associated with account information such as an ID and a password for each service. Moreover, you may be linked | related with residence information. Further, the information source table 123 may include other fields not shown. In this embodiment, account information, residence information, etc. are called registration information.

  The information acquisition unit 301 refers to the information source table 123 and acquires information from the network based on information for accessing each service. When the information acquisition application 121 functions as a news reader, the information acquisition unit 301 acquires the URI of the service whose service type corresponds to “news” from the information source table 123. Next, a news feed is acquired based on this URI.

  The acquired information storage 302 acquired by the information acquisition unit 301 is selected by the information selection unit 303 using the interest table 122 generated by the interest estimation unit 306 and the interest table display editing unit 307. FIG. 8 shows an example of the interest table 122.

  As shown in FIG. 8, the interest table 122 is associated with the degree of interest of the user for each category such as “politics” and “economy”. The interest level may include a set interest level set by the user using the interest table display editing unit 307 and an estimated interest level that is an interest level estimated by the interest estimation unit 306. The interest table 122 may be associated with information for calculating the estimated interest level. As an example of information for calculating the estimated degree of interest, FIG. 8 shows an index called an interest degree index that changes according to a user operation. The user operation is, for example, an operation for reading news. Details will be described later.

  As an example of a mode of selection by the information selection unit 303, for example, the information selection unit 303 may include a category in the acquired information storage 302 in which both the set interest level and the estimated interest level exceed a predetermined threshold in the interest table 122. It is like selecting only the news belonging to. Each news item belongs to one of the categories in the interest table 122.

  The all information acquisition unit 304 acquires more detailed information on the news selected by the information selection unit 303. The information to be acquired includes, for example, an image file. Further, when the RSS feed describes only a part of information, all information is acquired. The acquired information (all information storage 308) is displayed on the display device 14 by the information display means 305.

  FIG. 9 shows an example of a screen displayed on the display device 14 by the information display unit 305. As shown in FIG. 9A, the screen of the display device 14 displays, for example, a news title, an image of the news, and information such as a lead sentence and a text acquired from a news distribution service that distributes the news. Is done.

<Account aggregator>
Further, the information acquisition application 121 functions as an account aggregator by having the following configuration. The functional configuration is the same as that shown in FIG. In the information source table 123, registration information is associated with each information source (FIG. 7).

  The information acquisition unit 301 refers to the information source table 123 and acquires information from the network based on information for accessing each service and registration information. For example, when acquiring information from a service called SNSA, the information is acquired by accessing the URI of the SNSA and automatically logging in to the service using IDforA and PasswordforA. The structure and screen transition of the service website called SNSA must be analyzed in advance. The information acquisition unit 301 automatically acquires information based on the analysis result.

  The subsequent processing is the same as the processing in the news reader. However, the interest table 122 may be different from that of the news reader. In this case, for example, it is assumed that the user's friend includes information on a friend that the user wants to receive feed and article information and a friend that the user does not want to receive (see FIG. 10). In FIG. 10, the former set degree of interest is “1” and the latter degree of interest is “0”.

  As shown in FIG. 9B, on the screen of the display device 14, for example, a friend's remarks are displayed together with a friend's image.

<News display function and news clip function>
The screen displayed on the display device 14 includes various screens in addition to the screen illustrated in FIG. 9, and one of them is the screen illustrated in FIG. 11. FIG. 11 is an example of a screen displayed on the display device 14. The example shown in FIG. 11 is called a main screen as an example, and is a screen that serves as a hub when using the function provided by the information acquisition application 121. From this main screen, the user inputs an operation such as clicking on the tile 401 and uses functions such as a news reader and an account aggregator.

  When the tile 401 relates to news distributed by the news distribution service, the news screen as shown in FIG. 9A is displayed by the internal processing of the news reader as described above. When a part of the article is clicked or tapped, the information display unit 305 causes a context bar 406 as shown in FIG. 12 to appear. In the context bar 406, a clip button 404 and an SNS cooperation button 405 are displayed so as to be selectable.

  Here, when the user selects the clip button 404, the news displayed on the screen is clipped and stored as the clip information storage 309 shown in FIG. The specific storage destination of the clip information storage 309 may be the primary storage 11 or the secondary storage 12. The clip information storage 309 can be displayed as a list or displayed in detail by the information display unit 305.

<Presentation information selection process>
The technical idea described in the first embodiment is used, for example, when configuring the main screen of FIG. 11 in the present embodiment. On the main screen, for example, four of news articles distributed within 5 minutes are selected and displayed on the tile 401, and four of news articles distributed within 30 minutes are selected. Then, a process of displaying on the tile 401 is performed. The information selection unit 303 in FIG. 6 performs such a selection process, and performs the display process of the information selected by the information display unit 305.

<Interest estimation function>
The interest estimation means 306 monitors the operation input for using the clip function by the user as described above, and uses it for estimation of interest. As the monitoring operation, there are operations such as clipping from the list (tile 401) on the main screen, clipping from the detailed display of the news article (FIG. 12), and reading and displaying the clip information storage 309 again. For these operations, processing is performed to increase the interest index of the category to which the news that performed the operation belongs.

  Note that the interest estimation unit 306 is not limited to the operation input for using the clip function. For example, the user selects the SNS cooperation button 405 and writes an article of his / her SNS or displays a news article. A process of changing the interest index may be performed by monitoring an operation such as scrolling. In these operations, since it is estimated that the user is interested in the news article, processing for increasing the interest index is performed.

  Further, not only the above-described process having a positive correlation between the operation and the interest index but also a process having a negative correlation may be performed depending on the type of operation. That is, an operation in which the news displayed on the main screen is not read without being clicked or tapped at all is monitored, and the interest index for the category to which the news belongs is reduced.

  In addition, the act of displaying the news again after the user has clipped the news using the clip function is compared with the case where the news and the category to which the news belongs are compared to the case where the news is simply clipped and never viewed. It is estimated to be very interested. Therefore, for the act of displaying the news again after clipping it, a process of increasing the interest index relatively large is performed.

  By performing the above-described processing according to the present embodiment, it is possible to acquire a user's interest tendency with high accuracy. As a result, the accuracy of personalization increases.

  A more specific configuration of the above-described interest estimation function will be described below. In FIG. 13, the example of the change degree of the interest index by user operation in this embodiment is shown. The interest estimation means 306 holds a table as shown in FIG. 13 (or may refer to it). In FIG. 13, 1W, 1M, 3M, and 1Y mean one week, one month, three months, and one year, respectively, and represent the time that has elapsed since the user operation was performed. The number represents a coefficient to be added to the interest degree index.

  For example, a certain user clips a news article a belonging to the category “soccer” on the main screen three months ago, displays it again from the clip one week ago, and a news article b also belonging to “soccer” one week ago. Suppose you have a case where you have already read without opening it on the main screen. In this case, a value obtained by adding 1.15 × 1.4 × 0.5 to the interest index immediately before these user operations are performed is the changed interest index.

  FIG. 13 shows that the interest index may be increased and the interest index may be decreased depending on the type of user operation. As described above, in this embodiment, not only a user operation estimated to be interested in a sender of a certain category or information but also a user operation estimated to be not interested is monitored. Used for estimation. Therefore, according to this embodiment, the tendency of a user's interest can be estimated more accurately.

  Further, FIG. 13 shows that the effect of the user operation is attenuated according to the elapsed time from the time when the user operation is performed. In other words, it is shown that the degree of change in the degree of interest index change based on the user operation is set to decrease with time. This is to cope with the actual situation that the user's interest changes with time. In this way, by configuring so that the effect of the change in the interest index due to the user operation is attenuated, the information acquisition application 121 can appropriately cope with the future change in the interest of the user.

  In this embodiment, in order to configure the main screen, the information selection unit 303 in FIG. 6 performs the selection process described in the first embodiment, and the information display unit 305 selects the information selected. Perform display processing. For this reason, the presentation information that the user sees on the main screen or the like correctly reflects his / her interest, and may be interesting. According to the present embodiment, such effective information selection and display can be provided.

DESCRIPTION OF SYMBOLS 1 Information processing apparatus 2 Presentation candidate information acquisition means 3 Presentation candidate information evaluation means 4 Genre interest coefficient memory | storage means 5 Presentation information selection means 6 Presentation information evaluation means 100 Information processing apparatus 200 Server on cloud | cloud 121 121 Information acquisition application 122 Interest table 123 Information Source table 301 Information acquisition means 302 Acquisition information storage 303 Information selection means 304 All information acquisition means 305 Information display means 306 Interest estimation means 307 Interest table display editing means 308 All information storage 309 Clip information storage 3051 Registration information monitoring means 3052 Registration not yet completed Determination means 3053 Display control means 401 Tile 404 Clip button 405 SNS cooperation button 406 Context bar

Claims (8)

Presentation candidate information acquisition means for acquiring presentation candidate information associated with any of a plurality of genres;
A presentation candidate information evaluation means for evaluating the importance of each obtained presentation candidate information;
The degree of importance for each presentation candidate information is corrected based on the estimated value of the degree of interest of the user associated with each genre, and the probability of selecting the presentation candidate information as presentation information is calculated based on the degree of importance after correction. And presentation information selection means for selecting presentation information based on the probability,
An information processing apparatus comprising:   The presentation information selection means selects the certain presentation candidate information as presentation information based on the relationship between the importance after correction of all the presentation candidate information and the importance after correction of some of the presentation candidate information. The information processing apparatus according to claim 1, wherein a probability is calculated.   The information processing apparatus according to claim 2, wherein the probability is a ratio of the importance after correction of certain presentation candidate information to the total sum of importance after correction of all the presentation candidate information.   The presentation information selection means performs selection of presentation information one by one, and all the presentation candidate information is presentation candidate information that is a target to be acquired and a probability is calculated, and is not selected as presentation information The information processing apparatus according to claim 2, wherein the information processing apparatus is a device.   5. The method according to claim 1, further comprising a presentation information evaluation unit that changes an estimated value of a degree of interest in a genre associated with the presentation information in accordance with a user's reaction to the presentation information. The information processing apparatus described in 1. The presentation information evaluation means includes
Modify to increase or decrease the estimated value so far by at least browsing operations on content of the same genre presented in the past,
For the operation of the corrected content with high importance, the estimated value is corrected so that the interest level is increased small or the interest level is greatly decreased,
The information processing apparatus according to claim 5, wherein the estimated value is corrected so that the degree of interest greatly increases or the degree of interest decreases with respect to an operation of content with low importance. A presentation candidate information obtaining step for obtaining presentation candidate information associated with any of a plurality of genres;
A presentation candidate information evaluation step for evaluating the importance of each obtained presentation candidate information;
Correcting the importance for each presentation candidate information based on the estimated value of the degree of interest of the user associated with each genre;
Calculating a probability of selecting presentation candidate information as presentation information based on the importance after correction;
A presentation information selection step of selecting presentation information based on the probability;
An information processing method comprising: In the information processing device,
Presentation candidate information acquisition processing for acquiring presentation candidate information associated with any of a plurality of genres;
A presentation candidate information evaluation process for evaluating the importance of each obtained presentation candidate information;
A process of correcting the importance for each presentation candidate information based on the estimated value of the degree of interest of the user associated with each genre;
A process of calculating a probability of selecting the candidate presentation information as the presentation information based on the corrected importance,
A presentation information selection process for selecting presentation information based on the probability;
A program characterized by having executed.

Images