JP2017173912A - Correction apparatus, correction method, and correction program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To improve the accuracy of an evaluation value of a content.SOLUTION: An information processing apparatus includes an acquisition unit and a correction unit. The acquisition unit acquires a lapse time which is the time elapsed after a user takes action for a predetermined content. The correction unit corrects an evaluation value of a content to be distributed, based on the action, in accordance with the lapse time acquired by the acquisition unit.SELECTED DRAWING: Figure 4

Description

  The present invention relates to a correction apparatus, a correction method, and a correction program.

  In recent years, with the rapid spread of the Internet, advertisement distribution via the Internet has been actively performed. For example, advertising content such as companies and products is displayed in the advertising space set in the advertising medium (web page), and when such advertising content is clicked, advertising distribution is performed to transition to the advertiser's web page. ing.

  Here, for the advertisement to be evaluated, the behavior rate distribution obtained from the number of times the user took action and the number of times the advertisement was displayed was corrected using the past behavior rate, and the final value was calculated using the corrected value. A technique for calculating the evaluation value is proposed (Patent Document 1).

JP2013-161359A

  However, with the above-described conventional technology, it is not always possible to improve the accuracy of the evaluation value of content (for example, advertising content). Specifically, in the above-described conventional technology, the number of times of displaying the advertisement to be evaluated and the number of times the viewer has acted according to the advertisement are acquired, and the viewer is based on the acquired value. Create a probability distribution of the probability that will take action. And in the above-mentioned conventional technology, for each advertisement displayed in the past, the behavior rate that is the rate at which the viewer took action is acquired, the probability distribution is corrected according to the frequency of the acquired behavior rate, and corrected. Use the value to calculate the final ad rating.

  As described above, in the above-described conventional technology, the probability distribution calculated from the information obtained up to the present time is corrected using the action rate calculated from the past information. However, even if the past information is taken into account, the number is sufficient. Since the number of display times and the number of actions are not always obtained, the accuracy of the evaluation value of the content cannot always be improved.

  The present application has been made in view of the above, and an object thereof is to provide an information processing apparatus capable of increasing the accuracy of content evaluation values.

  The correction device according to the present application includes an acquisition unit that acquires an elapsed time since the user took an action related to a predetermined content, and the content of a distribution candidate based on the action according to the elapsed time acquired by the acquisition unit. And a correction unit that corrects the evaluation value.

  According to one aspect of the embodiment, there is an effect that the accuracy of the evaluation value of the content can be increased.

FIG. 1 is a diagram illustrating an example of information processing according to the embodiment. FIG. 2 is a diagram illustrating the transition of the conversion rate according to the elapsed time. FIG. 3 is a diagram illustrating a configuration example of the information processing system according to the embodiment. FIG. 4 is a diagram illustrating a configuration example of the information processing apparatus according to the embodiment. FIG. 5 is a diagram illustrating an example of a time information storage unit according to the embodiment. FIG. 6 is a diagram illustrating an example of the advertisement content storage unit according to the embodiment. FIG. 7 is a diagram illustrating an example of the weight information storage unit according to the embodiment. FIG. 8 is a flowchart illustrating an information processing procedure performed by the information processing apparatus according to the embodiment. FIG. 9 is a diagram illustrating the transition of the conversion rate according to the passage of time for each category to which the advertising content belongs. FIG. 10 is a diagram illustrating an example of the weight information storage unit according to the modification. FIG. 11 is a diagram illustrating an example of the advertising content storage unit according to the modification. FIG. 12 is a diagram illustrating an example of the model information storage unit according to the modification. FIG. 13 is a diagram illustrating an example of information processing according to the modification. FIG. 14 is a diagram illustrating an example of information processing according to the modification. FIG. 15 is a hardware configuration diagram illustrating an example of a computer that implements the functions of the information processing apparatus.

  Hereinafter, a mode for carrying out an information processing apparatus, an information processing method, and an information processing program according to the present application (hereinafter referred to as “embodiment”) will be described in detail with reference to the drawings. Note that the correction device, the correction method, and the correction program according to the present application are not limited by this embodiment.

[1. Information processing)
First, a series of information processing including correction processing according to the embodiment will be described with reference to FIG. 1 using an example. FIG. 1 is a diagram illustrating an example of information processing according to the embodiment. In the example of FIG. 1, the information processing system 1 includes a terminal device 10 and an information processing device 100. The terminal apparatus 10 and the information processing apparatus 100 are connected via a network so that they can communicate with each other by wire or wirelessly. Note that the information processing system 1 illustrated in FIG. 1 may include a plurality of terminal devices 10 and a plurality of information processing devices 100.

  An information processing apparatus 100, which is an example of a correction apparatus, is a server apparatus that distributes advertising content submitted by an advertiser. The information processing apparatus 100 according to the embodiment performs advertisement distribution called retargeting advertisement. The retargeting advertisement is, for example, a method of displaying advertisement content related to the predetermined website on the terminal device 10 of the user who has accessed the predetermined website designated by the advertiser.

  In addition, the information processing apparatus 100 according to the embodiment has a target value of cost per conversion (CPA) that is a cost required to acquire one result that leads to profits such as product purchase and member registration from advertisement content. By having the advertiser set a certain target cost per conversion, it has a function of calculating a bid set for the advertising content based on the target cost per conversion.

  Specifically, for each advertising content, the information processing apparatus 100 multiplies the conversion rate of the advertising content by the target conversion unit price set for the advertising content, and selects the advertising content. The bid price, which is the amount charged for, is calculated. Then, the information processing apparatus 100 performs advertisement distribution based on the calculated bid price.

  Here, the conversion rate is an actual value calculated by dividing the number of conversions from the advertising content (conversion number) by the number of clicks that the advertising content has been clicked so far. For this reason, it can be said that as the number of conversions and clicks increases, the accuracy of the conversion rate increases statistically. In other words, the conversion rate accuracy is statistically low when the number of conversions and clicks is small.

  That is, it can be said that the conversion rate is an unstable value that changes according to the performance information. For this reason, the conversion rate once calculated is not necessarily a certain value. Originally, it is desirable to use a sufficient number of clicks with sufficient reliability and a conversion rate calculated from the number of conversions. However, there are many advertisement contents that are difficult to click in the first place, and even if the advertisement contents are generally clicked. The probability of conversion is very low. Therefore, it may be difficult to accurately calculate the conversion rate for all advertisement contents.

  For example, when a bid price is calculated using a low-accuracy conversion rate, the calculated bid price is not always accurate, and therefore, advertisement distribution with high advertiser satisfaction may not be possible.

  Therefore, the information processing apparatus 100 according to the embodiment performs a correction process for correcting the conversion rate obtained from the record information or the like. Specifically, the information processing apparatus 100 uses the terminal device 10 to change the conversion rate (evaluation) of the advertising content corresponding to the predetermined content according to the elapsed time since the user took action related to the predetermined content. Example of value) is corrected.

  As described above, since the information processing apparatus 100 performs advertisement distribution by retargeting, in the present embodiment, the predetermined content indicates a predetermined website (web page) designated by the advertiser. . In the present embodiment, the action related to the predetermined content by the user indicates that the user accesses the predetermined website. Therefore, more specifically, the information processing apparatus 100 is an acquisition request based on an access to a predetermined website specified by the advertiser and then an access to the predetermined website. The conversion rate of the advertising content corresponding to the predetermined content is corrected according to the elapsed time elapsed until the content acquisition request is received.

  Here, the relationship between the elapsed time and the conversion rate of the advertising content will be described with reference to FIG. FIG. 2 is a diagram illustrating the transition of the conversion rate according to the elapsed time. Specifically, FIG. 2 shows performance information such as the number of times the advertisement content delivered to the terminal device 10 has been clicked by retargeting (the number of clicks) and the number of conversions from the clicked advertisement content (the number of conversions). The conversion rate calculated using is plotted against the elapsed time. The elapsed time here is from when the user accesses a predetermined website designated by the advertiser until the information processing apparatus 100 receives an acquisition request for advertising content related to the predetermined website. Is the elapsed time.

  As shown in FIG. 2, it is recognized that the conversion rate is higher as the elapsed time is shorter and tends to decrease as the elapsed time increases. The reason for this tendency is that while the elapsed time is short, the user's interest in the target is increasing, so the user can easily purchase the target, and as time passes It is conceivable that the interest will fade and it will be difficult for the user to purchase the target.

  As described above, the conversion rate is calculated by dividing the number of conversions by the number of clicks, and it can be said that the conversion rate calculated in this way is an average value. However, in practice, as shown in FIG. 2, the conversion rate changes according to the elapsed time, and therefore the accuracy can be improved by applying correction according to the elapsed time.

  Note that a value obtained by correcting a conversion rate (hereinafter referred to as “CVR”) simply calculated from the actual information without considering the elapsed time using the elapsed time is a predicted value, and thus the predicted value is CVR. The predicted conversion rate (hereinafter referred to as “predicted CVR”). Hereinafter, returning to FIG. 1, an example of information processing performed by the information processing apparatus 100 will be specifically described.

  First, it is assumed that the information processing apparatus 100 receives an advertisement content submission from an advertiser and information settings related to the advertisement content in advance and stores them in the advertisement content storage unit. FIG. 1 illustrates an example in which the information processing apparatus 100 receives, for example, the submission of the advertising content AD11 from the advertiser C1 and the settings of the distribution condition “P11” and the target CPA “15,000”. Note that the advertiser C1 submits a document using the advertiser device 20 described later.

  Here, the distribution condition is information set by the advertiser in order to perform the retargeting advertisement. For example, the advertiser sets the information processing apparatus 100 to provide the advertisement content submitted by the user who visited the web page by setting the web page managed by the advertiser as a delivery condition. . That is, in FIG. 1, an example in which the advertiser C1 instructs the information processing apparatus 100 to provide the advertisement content AD11 having contents related to the web page P11 to the user who visited the web page P11. Show. The advertiser C2 is also as shown in FIG.

  Here, it is assumed that the terminal device 10 accesses the web page P11 at “19:00 on March 13, 2016” in accordance with the operation of the user U1 (step S1). In the web page P11, a retage tag Tg15 is set as a retargeting tag (hereinafter referred to as “retag tag”).

  The retag tag is set on a web page or the like, for example, by an advertiser for retargeting advertisement, and causes the terminal device 10 to perform information notification called retargeting notification (hereinafter referred to as “retage notification”). As described above, since the web page P11 is managed by the advertiser C1, it is assumed that the retage tag Tg15 is set in advance on the web page P11 by the advertiser C1. In addition, the retage tag Tg15 provides the information processing apparatus 100 with the user ID, the page ID of the web page P11 including the retage tag Tg15, and the access date and time when the terminal apparatus 10 accesses the web page P11 as a retage notification. It shall be transmitted.

  That is, in FIG. 1, the terminal device 10 accesses the web page P11 in accordance with the operation of the user U1, so that the user ID “U1”, the page ID “P11”, and the access date “March 13, 2016, 19:00”. Is transmitted to the information processing apparatus 100 (step S2). Further, the information processing apparatus 100 stores the various information received in step S2 in the time information storage unit in association with each other.

  Further, it is assumed that the terminal device 10 accesses the web page P21 at “14:00 on March 15, 2016” in accordance with the operation of the user U1 (step S3). In the web page P21, a retage tag Tg16 is set. As described above, since the web page P12 is managed by the advertiser C2, it is assumed that the retage tag Tg16 is set in advance on the web page P21 by the advertiser C2. Similarly to the retage tag Tg15, the retage tag 16 uses the user ID, the page ID of the web page P21 including the retage tag Tg16, and the access date and time when the terminal device 10 accessed the web page P21 as a retage notification. Is transmitted to the information processing apparatus 100.

  In other words, in FIG. 1, the terminal device 10 accesses the web page P21 according to the operation of the user U1, so that the user ID “U1”, the page ID “P21”, and the access date “March 15, 2016, 14:00”. Is transmitted to the information processing apparatus 100 (step S4). Further, the information processing apparatus 100 stores the various information received in step S3 in the time information storage unit in association with each other.

  Here, it is assumed that the terminal device 10 transmits an advertisement content acquisition request to the information processing apparatus 100 based on the access to the web pages P11 and P21 according to the operation of the user U1 (step S5). The advertisement content acquisition request based on the access to the web pages P11 and P21 includes, for example, an advertisement space AD11 corresponding to the web page P11 and an advertisement frame capable of displaying the advertisement content AD21 corresponding to the web page P21. This is an acquisition request for accessing a predetermined content. The acquisition request includes the user ID “U1” of the terminal device 10.

  When receiving the advertisement content acquisition request from the terminal device 10, the information processing apparatus 100 acquires the request reception date and time when the acquisition request is received (step S6). For example, the information processing apparatus 100 acquires the request reception date and time from a predetermined timer in the own apparatus. Here, it is assumed that the information processing apparatus 100 has acquired the request reception date and time “March 15, 2016, 22:00”.

  Further, the information processing apparatus 100 specifies advertisement content as a distribution candidate (step S7). For example, the information processing apparatus 100 refers to the time information storage unit and specifies that the web pages accessed so far by the user U1 of the acquisition request transmission source are the web pages P11 and P21. Next, the information processing apparatus 100 refers to the advertisement content storage unit, and since the web pages P11 and P21 are set as distribution conditions, the information processing apparatus 100 corresponds to the advertisement content AD11 corresponding to the web page P11 and the web page P21. The advertisement content AD21 to be identified is specified as the distribution candidate advertisement content.

  Next, the information processing apparatus 100 acquires the elapsed time that has elapsed since the user U1 accessed each of the web pages P11 and P21 until receiving the advertisement content acquisition request from the terminal apparatus 10 of the user U1 ( Step S8). For example, the information processing apparatus 100 acquires the difference between the access date and time and the request reception date and time as the elapsed time.

  In the example of FIG. 1, the information processing apparatus 100 acquires “53 hours” as the elapsed time from when the user U1 accesses the web page P11 until the acquisition request is received. Further, the information processing apparatus 100 acquires “8 hours” as the elapsed time from when the user U1 accesses the web page P21 until the acquisition request is received.

  Next, the information processing apparatus 100 corrects the CVR of each of the distribution candidate advertisement contents AD11 and AD21 according to the elapsed time acquired in Step S8 (Step S9). Specifically, the information processing apparatus 100 corrects the CVR of each of the distribution candidate advertisement contents AD11 and AD21 using the corresponding elapsed time.

  Here, referring to FIG. 2, it can be seen that the conversion rate is very high within the elapsed time “25 hours”. For this reason, for example, within the elapsed time “25 hours”, it is expected that the actual conversion rate will be higher than the CVR simply calculated from the performance information without considering the elapsed time. Therefore, the information processing apparatus 100 performs correction to greatly increase the CVR “0.03” of the advertisement content AD21 corresponding to the elapsed time “8 hours”.

  For example, the information processing apparatus 100 multiplies the CVR “0.03” of the advertising content AD21 by a predetermined weight value (here, “3”) to set the CVR of the advertising content AD21 to “0.09”. And correct. This corresponds to the fact that the information processing apparatus 100 predicts that the conversion rate is actually “0.09” at the elapsed time “8 hours” in the advertisement content AD21.

  Referring to FIG. 2, after the elapsed time “25 hours”, the CVR shows a relatively gradual decrease or can be regarded as constant. For this reason, for example, after the elapsed time “25 hours”, it is expected that the CVR simply calculated from the performance information without considering the elapsed time and the actual conversion rate are not significantly different. Therefore, the information processing apparatus 100 performs a correction for slightly increasing the CVR “0.05” of the advertisement content AD11 corresponding to the elapsed time “53 hours” or does not correct it.

  For example, the information processing apparatus 100 multiplies the CVR “0.05” of the advertising content AD11 by a predetermined weight value (here, “1.2”) to set the CVR of the advertising content AD11 to “0.06”. To "". This corresponds to the fact that the information processing apparatus 100 predicts that the conversion rate is actually “0.06” for the elapsed time “53 hours” in the advertisement content AD11.

  Next, the information processing apparatus 100 extracts advertisement content to be distributed (step S10). Specifically, the information processing apparatus 100 calculates an expected return value (eCPM: effective cost per mille) of each of the distribution candidate advertisement contents AD11 and AD21 based on the CVR corrected in step S9, that is, the predicted CVR. More specifically, the information processing apparatus 100 calculates a bid price (CPC: Cost Per Click) by multiplying the predicted CVR and the target CPA. Then, the information processing apparatus 100 calculates eCPM by multiplying the calculated CPC and a click rate (CTR: Click through Rate).

  Then, the information processing apparatus 100 extracts the advertising content with the higher calculated eCPM as the advertising content to be distributed among the advertising content AD11 and AD21. Then, the information processing apparatus 100 distributes the extracted advertisement content to the terminal apparatus 10 that is the acquisition request transmission source.

  As described above, the information processing apparatus 100 according to the embodiment corrects the conversion rate of the advertising content corresponding to the predetermined content according to the elapsed time since the user accessed the predetermined content. Thereby, since the information processing apparatus 100 can correct the conversion rate calculated from the actual value according to the conversion rate that changes according to the elapsed time, the conversion rate calculated from the actual value is a more accurate value. Can be predicted.

  Further, the information processing apparatus 100 according to the embodiment can accurately calculate the bid price by using the corrected conversion rate. Further, since the information processing apparatus 100 automatically sets the bid price calculated as described above on behalf of the advertiser by conversion optimization and performs the advertisement distribution, the information distribution apparatus 100 can perform the advertisement distribution with high advertiser satisfaction. .

[2. Configuration of information processing system]
Next, the configuration of the information processing system according to the embodiment will be described with reference to FIG. FIG. 3 is a diagram illustrating a configuration example of the information processing system 1 according to the embodiment. As illustrated in FIG. 3, the information processing system 1 includes a terminal device 10, advertiser apparatuses 20 _{1 to} 20 _n , an information providing apparatus 30, and an information processing apparatus 100. The terminal device 10, the advertiser devices 20 _{1 to} 20 _n , the information providing device 30, and the information processing device 100 are connected to be communicable via a network N by wire or wireless. Note that the information processing system 1 illustrated in FIG. 3 may include a plurality of terminal devices 10, a plurality of information providing devices 30, and a plurality of information processing devices 100.

  The terminal device 10 is a terminal device used by a user. The terminal device 10 is, for example, a smartphone, a tablet terminal, a notebook PC (Personal Computer), a desktop PC, a mobile phone, a PDA (Personal Digital Assistant), or the like. For example, the terminal device 10 obtains a web page from the information providing device 30 by accessing the information providing device 30 and displays the obtained web page on a display device (for example, a liquid crystal display). Further, when the web page includes an advertising space, the terminal device 10 acquires the advertising content from the information processing device 100 by accessing the information processing device 100, and displays the acquired advertising content on the web page. To do. However, the present invention is not limited to this example, and the terminal device 10 may acquire a web page including advertisement content from the information providing device 30. In this case, the information providing apparatus 30 distributes to the terminal apparatus 10 a web page in which the advertisement content distributed by the information processing apparatus 100 is incorporated.

Advertiser devices 20 _{1 to} 20 _n are terminal devices used by advertisers. The advertiser apparatuses 20 _{1 to} 20 _n perform advertisement content submission and information setting related to the advertisement content to the information processing apparatus 100 in accordance with the operations of the advertiser. The advertiser apparatuses 20 _{1 to} 20 _{n according} to the embodiment submit corresponding advertisement contents such as still images, moving images, and text data to the information processing apparatus 100.

In addition, the advertiser apparatuses 20 _{1 to} 20 _n specify predetermined web pages (for example, web pages managed by the advertiser) as distribution conditions, so that users who have visited such web pages can be provided. The information processing apparatus 100 is instructed to provide its own advertising content. The information processing apparatus 100 performs the retargeting advertisement by such setting by the advertiser.

In addition, the advertiser apparatuses 20 _{1 to} 20 _n set the target CPA as the advertising budget, thereby instructing the information processing apparatus 100 to perform advertisement distribution that can achieve the target CPA. The information processing apparatus 100 automatically calculates the bid price by performing such a setting for the advertiser by the conversion optimization function, and distributes the advertisement using the calculated bid price.

Also, the advertiser using the advertiser device 20 ₁ to 20 _n, without traffic advertising content to the information processing apparatus 100, in some cases to ask the agency trafficking advertising content. In this case, it is the agency that submits the advertising content to the information processing apparatus 100. In the following, the term “advertiser” is a concept that includes not only the advertiser but also the agency, and the term “advertiser device” includes the agency device used by the agency as well as the advertiser device. It is assumed to be a concept. In addition, since the advertiser devices 20 _{1 to} 20 _n have the same functions, respectively, in the following, when it is not necessary to distinguish the advertiser devices 20 _{1 to} 20 _n , they are collectively referred to as “advertiser device”. 20 ”.

  The information providing device 30 is a web server or the like that provides a web page to the terminal device 10. The information processing apparatus 30 is, for example, various web pages related to news sites, auction sites, weather forecast sites, shopping sites, finance (stock price) sites, route search sites, map providing sites, travel sites, restaurant introduction sites, web blogs, etc. I will provide a.

  The information processing apparatus 100 is a server apparatus that distributes the advertising content submitted from the advertiser apparatus 20 based on setting information set by the advertiser. As described above, the information processing apparatus 100 distributes the advertising content to the terminal apparatus 10 when accessed from the terminal apparatus 10. Further, when accessed from the information providing apparatus 30, the information processing apparatus 100 distributes advertisement content to the information providing apparatus 30.

[3. Configuration of information processing apparatus]
Next, the information processing apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 4 is a diagram illustrating a configuration example of the information processing apparatus 100 according to the embodiment. As illustrated in FIG. 4, the information processing apparatus 100 includes a communication unit 110, a storage unit 120, and a control unit 130.

(About the communication unit 110)
The communication unit 110 is realized by, for example, a NIC (Network Interface Card). The communication unit 110 is connected to the network by wire or wirelessly, and transmits and receives information to and from the terminal device 30, for example.

(About the storage unit 120)
The storage unit 120 is realized by, for example, a semiconductor memory device such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. The storage unit 120 includes a time information storage unit 121, an advertisement content storage unit 122-1, and a weight information storage unit 123-1.

(About time information storage unit 121)
The time information storage unit 121 stores an access date and time when a user accesses predetermined content. In the present embodiment, the time information storage unit 121 stores the access date and time when the user accesses the web page in which the retage tag is set, that is, the web page related to the retargeting advertisement as shown in FIG. Here, FIG. 5 shows an example of the time information storage unit 121 according to the embodiment. In the example of FIG. 5, the time information storage unit 121 includes items such as “user ID”, “web page ID”, and “access date / time”. The time information storage unit 121 will be described in more detail than the time information storage unit shown in FIG.

  “User ID” indicates identification information for identifying the user or the terminal device 10 of the user. “Web page ID” indicates identification information for identifying a web page. The “web page ID” only needs to identify a web page, and may be a URL (Uniform Resource Locator) of the web page, for example. The access date / time indicates the access date / time when the user accesses the corresponding web page, and corresponds to, for example, a combination of “year / month / day / time”. The access date and time is acquired by the acquisition unit 133 described later.

  That is, FIG. 5 shows an example in which the user identified by the user ID “U1” accesses the web page identified by the web page ID “P1” at “9:00 on March 13, 2016”. As shown in FIG. 1, the user ID may be used to distinguish between the users. For example, a user identified by the user ID “U1” is represented as a user U1. Similarly, the advertiser and advertisement content may be described using identification information.

(Advertising content storage unit 122-1)
The advertising content storage unit 122-1 stores various types of information related to the advertising content submitted from the advertiser device 20 and the advertising content accepted from the advertiser device 20. Further, the advertisement content storage unit 122-1 stores performance information related to advertisement distribution. Here, FIG. 6 shows an example of the advertisement content storage unit 122-1 according to the embodiment. In the example of FIG. 6, the advertisement content storage unit 122-1 includes “advertiser ID”, “advertisement ID”, “advertisement content”, “distribution condition”, “target CPA”, “number of clicks”, “CTR”, It has an item such as “number of conversions”. Note that the advertisement content storage unit 122-1 illustrated in FIG. 5 describes the advertisement content storage unit illustrated in FIG. 1 in more detail.

  “Advertiser ID” indicates identification information for identifying an advertiser or an advertiser terminal. “Advertisement ID” indicates identification information for identifying advertisement content. “Advertising content” indicates the advertising content submitted from the advertiser terminal 20. FIG. 6 shows an example in which conceptual information is stored in “advertising content”, but actually, still images, moving images, text data, or file path names indicating the storage locations thereof are stored. The

  The “distribution condition” is information set by the advertiser for performing the retargeting advertisement. For example, the advertiser instructs the information processing apparatus 100 to provide the advertisement content submitted by the user visiting the web page by setting the web page managed by the advertiser as a delivery condition. . That is, the “distribution condition” is condition information for determining what kind of advertisement content is distributed to the user in the retargeting advertisement.

  “Target CPA” indicates a result that leads to profits such as product purchase and member registration from the corresponding advertising content, that is, a target value of the conversion unit price, which is a cost for acquiring one conversion. In the conversion optimization, it is determined that the “target CPA” can be set only for the advertising content in which a predetermined number (for example, “10”) or more conversions are obtained.

  “Number of clicks” indicates the number of times the advertisement content has been clicked so far. “CTR” is a value obtained by dividing “number of clicks” by the number of times the advertisement content is displayed. The CTR of advertising content that has not been distributed to the terminal device 10 includes a predetermined fixed value, an average value of CTR for all advertising content, and the same advertising category (for example, “car”, “ The average value of CTR and the like in all advertisement contents belonging to “travel”) are stored. “Conversions” indicates the number of clicks on the advertising content that have been converted so far.

  That is, the example of FIG. 6 shows an example in which the advertiser identified by the advertiser ID “C1” submits the data “XX11” of the advertisement content as the advertisement content identified by the advertisement ID “AD11”. . In addition, an example is shown in which the advertiser C1 instructs the user who visited the web page “P11” to be routed to provide the advertisement content AD11. Further, an example is shown in which the advertiser C1 sets the target CPA “15,000” for the advertising content AD11.

(About the weight information storage unit 123-1)
The weight information storage unit 123-1 stores a weight value for correcting the CVR. For example, the weight information storage unit 123-1 stores a weight value based on a change in conversion rate according to the passage of time shown in FIG. For example, as shown in FIG. 2, the weight information storage unit 123-1 stores a higher weight value as the elapsed time is shorter, based on a tendency that the conversion rate is higher as the elapsed time is shorter. Here, FIG. 7 shows an example of the weight information storage unit 123-1 according to the embodiment. In the example of FIG. 7, the weight information storage unit 123-1 has items such as “elapsed time” and “weight value”.

  The “elapsed time” is an acquisition request based on accessing the predetermined content after the user accesses the predetermined content specified by the advertiser, and the information processing apparatus 100 sends the acquisition request for the advertising content. Indicates the elapsed time elapsed until reception. The “weight value” indicates a weight value based on the transition of the conversion rate according to the elapsed time as described above. In the present embodiment, the elapsed time is shown in unit time (1 hour), but is not limited thereto, and may be shown in units of seconds, minutes, days, or the like.

  Here, the setting of the weight value will be described more specifically. For example, referring to FIG. 2, it can be seen that the conversion rate is very high within the elapsed time “25 hours”. For this reason, for example, within the elapsed time “25 hours”, it is expected that the actual conversion rate will be higher than the CVR simply calculated from the performance information without considering the elapsed time. Therefore, within the elapsed time “25 hours”, a weight value for increasing the degree of correction as the elapsed time is shorter, and a weight value for correcting the CVR to a higher value is set.

  On the other hand, after the elapsed time “25 hours”, the conversion rate shows a relatively gradual decrease or can be regarded as constant. For this reason, for example, after the elapsed time “25 hours”, it is expected that the CVR simply calculated from the performance information without considering the elapsed time and the actual conversion rate are not significantly different. For this reason, after the elapsed time “25 hours”, in order to correct the CVR to a slightly higher value in response to a gradual decrease, the degree of correction is smaller than that within the elapsed time “25 hours”. A weight value is set. Also, a weight value that does not correct CVR, that is, a weight value “1” may be set.

  As an example, the straight line model M1 shown in FIG. 2 is an approximate curve calculated based on the variance of the plot, and a weight value is set based on the straight line model M1.

  That is, the example of FIG. 7 shows an example in which correction is made using the weight value “3.0” when the elapsed time is within “0 to 5 hours”.

(About the control unit 130)
Returning to FIG. 2, the control unit 130 executes various programs stored in a storage device inside the information processing apparatus 100 using a RAM as a work area by a CPU (Central Processing Unit), an MPU (Micro Processing Unit), or the like. Is realized. The control unit 130 is realized by an integrated circuit such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

  As shown in FIG. 4, the control unit 130 includes a reception unit 131, a reception unit 132, an acquisition unit 133, a correction unit 134, an extraction unit 135, and a distribution unit 136. Information described below Implement or execute a processing function or action. Note that the internal configuration of the control unit 130 is not limited to the configuration illustrated in FIG. 4, and may be another configuration as long as the information processing described below is performed. In addition, the connection relationship between the processing units included in the control unit 130 is not limited to the connection relationship illustrated in FIG. 4, and may be another connection relationship.

(About the reception unit 131)
The accepting unit 131 accepts the submission of the advertising content from the advertiser terminal 20 and the setting of various information regarding the advertising content, and stores the accepted advertising content and setting information in the advertising content storage unit 122-1. Specifically, the accepting unit 131 accepts submission of advertising content corresponding to still images, moving images, text data, etc., and settings of delivery conditions, target CPA, and the like.

  And the reception part 131 pays out advertiser ID and advertisement ID, matches the received advertiser ID and advertisement ID, and stores the information received as mentioned above in the advertisement content storage part 122-1.

(Receiver 132)
The receiving unit 132 receives an advertisement content acquisition request from the terminal device 10. For example, the receiving unit 132 receives an advertisement content acquisition request based on access to a predetermined content by a user who has transmitted the acquisition request. The acquisition request here is, for example, the fact that the user of the acquisition request transmission source has accessed the content specified by the advertiser who has set the retargeting advertisement in the past, so that the advertising content corresponding to the content is Indicates an acquisition request for acquisition.

  In addition, the reception unit 132 notifies the acquisition unit 133 of information indicating that the advertisement content acquisition request has been received.

(About the acquisition unit 133)
When the acquisition unit 133 receives the advertisement content acquisition request by the reception unit 132, the acquisition unit 133 acquires an elapsed time since the user of the acquisition request transmission source took an action related to the predetermined content. Specifically, the acquisition unit 133 acquires the elapsed time since the acquisition request transmission source user accessed the predetermined content.

  For example, the acquisition unit 133 acquires an access date and time when the user who has transmitted the acquisition request accesses predetermined content using the terminal device 10. In addition, the acquisition unit 133 acquires a request reception date and time that is a date and time when the reception unit 132 has received an acquisition request for advertisement content based on a user who has acquired the acquisition request transmitting the predetermined content. In this way, the acquisition unit 133 acquires advertisement content based on the fact that the predetermined request content is accessed by the acquisition request transmission source user after the acquisition unit 133 accesses the predetermined content. Get the elapsed time that elapses before the request is received. That is, the acquisition unit 133 acquires a difference time obtained by subtracting the access date and time from the request reception date and time as the elapsed time. The acquisition unit 133 acquires such elapsed time for each predetermined content accessed by the user.

  In the present embodiment, the predetermined content is content (web page) related to the retargeting advertisement as described above, for example, content managed by the advertiser. Below, an example of the acquisition process by the acquisition part 133 is demonstrated.

  For example, it is assumed that the terminal device 10 accesses the web page P11 at “19:00 on March 13, 2016” in accordance with the operation of the user U1. As shown in FIG. 1, since the retage tag Tg15 is set in the web page P11, the terminal device 10 has the user ID “U1”, the page ID “P11”, the access date and time “March 13, 2016”. 19:00 "is transmitted in association with the information processing apparatus 100. As a result, the acquisition unit 133 acquires the access date “March 13, 2016, 19:00” from the user U1. Then, the acquisition unit 133 stores the acquired user ID “U1”, page ID “P11”, and access date “March 13, 2016, 19:00” in the time information storage unit 121. The access date / time is referred to as access date / time T11.

  For example, it is assumed that the terminal device 10 accesses the web page P21 at “14:00 on March 15, 2016” in accordance with the operation of the user U1. As shown in FIG. 1, since the retage tag Tg16 is set in the web page P21, the terminal device 10 has the user ID “U1”, the page ID “P21”, and the access date “March 15, 2016. 14:00 ”is transmitted to the information processing apparatus 100 in association with each other. Thereby, the acquisition unit 133 acquires the access date “March 15, 2016, 14:00” from the user U1. Then, the acquiring unit 133 stores the acquired user ID “U1”, page ID “P21”, and access date “March 15, 2016, 14:00” in the time information storage unit 121. The access date / time is referred to as access date / time T21.

  Here, the access date and time will be described. So far, the access date and time has been described as the date and time when the terminal device 10 accessed the content in accordance with the operation of the user U1. However, the access date and time is not limited to this example. For example, after accessing the content, the user may browse the accessed content for a while. Therefore, the access date and time may be any time as long as the terminal device 10 is displaying the content. The access date / time may be a date / time closed due to a transition from the content to another content. The timing of the access date and time as described above may be set in advance, for example, by the advertiser controlling the retag tag.

  However, if the access date / time timing setting varies depending on the advertiser, inequality may occur in the calculated elapsed time. For this reason, it is desirable to unify in advance the timing of the access date and time.

  Returning to the description of the acquisition unit 133. Here, it is assumed that the terminal apparatus 10 transmits an advertisement content acquisition request to the information processing apparatus 100 based on the access to the web pages P11 and P21 according to the operation of the user U1. The acquisition request includes the user ID “U1”.

  When receiving the advertisement content acquisition request from the terminal device 10 of the user U1, the reception unit 132 notifies the acquisition unit 133 of, for example, the user ID “U1” as information indicating that the acquisition request has been received. Upon receiving the user ID “U1” from the reception unit 132, the acquisition unit 133 acquires the current date and time as the request acquisition date and time. For example, the acquisition unit 133 acquires the request reception date and time from a predetermined timer in the information processing apparatus 100. Here, it is assumed that the acquisition unit 133 acquires the request reception date and time “March 15, 2016, 22:00”. The acquisition request is referred to as acquisition request AR1, and the request reception date and time is referred to as request reception date and time T1.

  Note that the request acquisition date and time may be acquired by the reception unit 132. For example, when receiving the advertisement content acquisition request, the reception unit 132 acquires the current date and time as a request reception date and time from a predetermined timer in the information processing apparatus 100. Then, the reception unit 132 associates the acquired request reception date and time with the user ID and notifies the acquisition unit 133 of the association.

  Next, the acquisition unit 133 specifies advertisement content as a distribution candidate. For example, the acquisition unit 133 refers to the time information storage unit 121 illustrated in FIG. 5 and specifies that the web pages accessed by the user U1 so far are the web pages P11 and P21. In the example of FIG. 5, the user U1 accesses the web page P31 in addition to the web pages P11 and P21.

  Here, in the information processing apparatus 100, in the retargeting advertisement, the advertisement content to be distributed is set to be the advertisement content corresponding to the web page accessed within the past three months from the current time when the advertisement content acquisition request is received. It is assumed that In the example of FIG. 5, the user U1 accesses the web page P31 at “10:25 on October 16, 2015”, and the request reception date and time “March 15, 2016” at which the acquisition request AR1 was received. It is more than 3 months before 22:00. For this reason, the acquisition part 133 specifies that the web pages which the user U1 accessed so far are the web pages P11 and P21.

  Further, the acquisition unit 133 refers to the advertisement content storage unit 122-1 illustrated in FIG. 6, and the web pages P <b> 11 and P <b> 21 are set as distribution conditions. Therefore, the advertisement content AD <b> 11 corresponding to the web page P <b> 11, and The advertisement content AD21 corresponding to the web page P21 is specified as the distribution candidate advertisement content. It should be noted that such a specifying process for specifying the advertisement content of the distribution candidate may be performed by a processing unit (for example, a specifying unit) other than the acquisition unit 133.

  And the acquisition part 133 acquires the elapsed time passed after the user U1 accessed each of the web pages P11 and P21 until it receives the acquisition request AR1.

  Specifically, the acquisition unit 133 acquires the elapsed time that has elapsed since the user U1 accessed the web page P11 until the acquisition request AR1 was received from the terminal device 10 of the user U1. In addition, the acquisition unit 133 acquires the elapsed time that has elapsed since the user U1 accessed the web page P21 until the acquisition request AR1 is received from the terminal device 10 of the user U1.

  That is, the acquisition unit 133 subtracts the access date / time T11 “March 13, 2016 19:00” from the request reception date T1 “March 15, 2016 22:00” when the acquisition request AR1 was received. "53 hours" is acquired as the elapsed time elapsed from when the user U1 accesses the web page P11 until the acquisition request AR1 is received.

  Further, the acquisition unit 133 subtracts the access date / time T12 “March 15, 2016 14:00” from the request reception date T1 “March 15, 2016 22:00” when the acquisition request AR1 was received. "8 hours" is acquired as the elapsed time elapsed from when the user U1 accesses the web page P21 until the acquisition request AR1 is received.

  Now, although the acquisition part 133 showed the example which acquires the request | requirement reception date and time when the acquisition request of advertisement content was received by the receiving part 132 so far, it is not limited to this example. For example, the acquisition unit 133 may acquire the request transmission date and time when the terminal device 10 transmits the advertisement content acquisition request. In such a case, the terminal device 10 transmits an acquisition request including the user ID and the current time as the request transmission date and time to the information processing device 100.

(About the correction unit 134)
The correction unit 134 corrects the evaluation value of the distribution candidate content according to the elapsed time acquired by the acquisition unit 133. Specifically, the correction unit 134 corrects the evaluation value of the distribution candidate content that is advertisement content corresponding to the predetermined content accessed by the acquisition request transmission source user. For example, the correction unit 134 uses the weight value according to the elapsed time acquired by the acquisition unit 133, and is an advertisement content corresponding to a predetermined content accessed by the user of the acquisition request transmission source. The CVR that is the evaluation value of the content is corrected. At this time, for example, the correction unit 134 corrects the CVR using a higher weight value as the elapsed time acquired by the acquisition unit 133 is shorter. Hereinafter, an example of correction processing by the correction unit 134 will be described.

  Here, as described above, it is assumed that the acquisition unit 133 has acquired the elapsed time “53 hours” corresponding to the web page P11 and the elapsed time “8 hours” corresponding to the web page P21. For example, the correction unit 134 refers to the weight information storage unit 123-1 illustrated in FIG. 7, and uses the weight values corresponding to the elapsed times acquired by the acquisition unit 133, and the advertising content corresponding to the web page P11. The CVR of each of the advertising content AD21 corresponding to the AD11 and the web page P21 is corrected.

  For the advertising content AD11, first, the correction unit 134 refers to the advertising content storage unit 122-1, and calculates the CVR. In the example of the advertisement content storage unit 122-1 illustrated in FIG. 6, since the number of clicks “3,000” and the number of conversions “150” of the advertisement content AD11 are corrected, the correction unit 134 By dividing by “3,000”, the CVR “0.05” of the advertising content AD11 is calculated.

  Next, the correction unit 134 refers to the weight information storage unit 123-1 illustrated in FIG. 7, and specifies the weight value “1.2” for the elapsed time “53 hours” corresponding to the advertisement content AD11. Then, the correcting unit 134 corrects the CVR of the advertising content AD11 to “0.06” by multiplying the CVR “0.05” by the weight value “1.2”. This corresponds to the prediction that the correction unit 134 actually predicted that the conversion rate would be “0.06” in the advertisement content AD11 at the elapsed time “53 hours”.

  Next, for the advertising content AD21, the correction unit 134 refers to the advertising content storage unit 122-1, and calculates the CVR. In the example of the advertising content storage unit 122-1 illustrated in FIG. 6, since the number of clicks “9,000” and the number of conversions “270” of the advertising content AD21, the correction unit 133 selects the number of clicks “270” of the conversion number “270”. By dividing by “9,000”, CVR “0.03” of the advertising content AD21 is calculated.

  Next, the correction unit 134 refers to the weight information storage unit 123 illustrated in FIG. 7 and specifies the weight value “2.8” for the elapsed time “8 hours” corresponding to the advertisement content AD21. Then, the correcting unit 134 corrects the CVR of the advertisement content AD21 to “0.08” by multiplying the CVR “0.03” by the weight value “2.8”. This corresponds to the fact that the correction unit 134 has predicted that the conversion rate is actually “0.08” at the elapsed time “8 hours” in the advertisement content AD21.

(About the extraction unit 135)
The extraction unit 135 extracts advertisement content to be distributed. Specifically, the extraction unit 135 extracts advertisement content to be distributed based on the predicted CVR that is the corrected CVR corrected by the correction unit 134. More specifically, the extraction unit 135 calculates CPC by multiplying the predicted CVR and the target CPA. Then, the extracting unit 135 extracts advertisement content to be distributed based on the eCPM obtained by multiplying the calculated CPC and CTR. Hereinafter, an example of correction processing by the extraction unit 135 will be described.

  Here, as described above, it is assumed that the correction unit 134 has calculated the predicted CVR “0.06” of the distribution candidate advertisement content AD11 and the predicted CVR “0.08” of the distribution candidate advertisement content AD21. .

  For the advertising content AD11, first, the extraction unit 135 refers to the advertising content storage unit 122-1, and calculates the CPC. In the example of the advertising content storage unit 122-1 illustrated in FIG. 6, the target CPA of the advertising content AD <b> 11 is “15,000”. Therefore, the extraction unit 135 calculates the CPC “900” by multiplying the predicted CVR “0.06” by the target CPA “15,000”.

  In addition, the extraction unit 135 refers to the advertisement content storage unit 122-1 to calculate eCPM. In the example of the advertisement content storage unit 122-1 illustrated in FIG. 6, the CTR of the advertisement content AD11 is “0.05”. Therefore, the extraction unit 135 calculates eCPM “45” by multiplying CPC “900” by CTR “0.05”.

  Next, for the advertising content AD21, in the example of the advertising content storage unit 122-1 illustrated in FIG. 6, the target CPA of the advertising content AD21 is “20,000”. Therefore, the extraction unit 135 calculates the CPC “1,600” by multiplying the predicted CVR “0.08” by the target CPA “20,000”. In the example of the advertising content storage unit 122-1 illustrated in FIG. 6, the CTR of the advertising content AD21 is “0.03”. Therefore, the extraction unit 135 calculates eCPM “48” by multiplying CPC “1,600” by CTR “0.03”.

  Then, the extraction unit 135 extracts, from the advertisement content storage unit 122-1, the advertisement content AD21 having a high eCPM among the advertisement content AD11 and AD21 that are distribution candidates.

(About distribution unit 136)
The distribution unit 136 distributes advertising content. Specifically, the distribution unit 136 distributes the advertising content extracted by the extraction unit 135 to the terminal device 10. For example, it is assumed that the advertisement content AD21 is extracted by the extraction unit 135. In this case, the distribution unit 136 attaches predetermined link information (for example, URL of the landing page) to the advertisement content AD21 and distributes the information including the information indicating the CPC “1,600” to the terminal device 10.

  Further, the distribution unit 136 may accept click information that is information indicating that the advertisement content has been clicked from the terminal device 10. For example, every time the advertisement content AD21 is clicked, the terminal device 10 transmits click information including the advertisement ID “AD21”, the number of clicks “1”, and the CPC “1600” to the information processing apparatus 100. When receiving the click information, the distribution unit 136 adds “1” to the number of clicks corresponding to the advertisement content AD21 in the advertisement content storage unit 122-1. Although not shown, the distribution unit 136 adds “1,600” to the current billing amount corresponding to the advertisement content AD21.

[4. Processing procedure)
Next, an information processing procedure executed by the information processing apparatus 100 according to the embodiment will be described with reference to FIG. FIG. 8 is a flowchart illustrating an information processing procedure performed by the information processing apparatus 100 according to the embodiment.

  First, in the information processing apparatus 100, the acquisition unit 133 determines whether or not the user has acquired the access date and time when the user accessed the predetermined web page related to the targeting advertisement using the terminal device 10 (step S101). If the access date / time cannot be acquired (step S101; No), the acquisition unit 133 waits until it can be acquired.

  On the other hand, when the access date and time is acquired by the acquisition unit 133 (step S101; Yes), the reception unit 132 determines whether or not an advertisement content acquisition request has been received (step S102). If the acquisition unit 132 has not received an acquisition request (step S102; No), the reception unit 132 waits until an acquisition request is received.

  On the other hand, when the acquisition request is received by the reception unit 132 (step S102; Yes), the acquisition unit 133 acquires a request reception date and time that is the date and time when the acquisition request is received (step S103). Next, the acquisition unit 133 specifies advertisement content as a distribution candidate (step S104).

  And the acquisition part 133 acquires the elapsed time after the user of an acquisition request transmission source accessed the web page relevant to the said advertisement content about each advertisement content of the specified delivery candidate (step S105). Specifically, the acquisition unit 133 acquires the elapsed time from when the acquisition request transmission source user accessed the web page until the reception unit 132 receives the advertisement content acquisition request. That is, the acquisition unit 133 acquires a difference time obtained by subtracting the access date and time from the request reception date and time as the elapsed time.

  Next, the correction unit 134 corrects the CVR of each of the distribution candidate advertisement contents based on the elapsed time acquired by the acquisition unit 133 (step S106). Specifically, the correction unit 134 performs correction according to the elapsed time acquired by the acquisition unit 133 for each of the CVRs of the distribution candidate advertisement contents. More specifically, the correcting unit 134 corrects the weighting value according to the elapsed time acquired by the acquiring unit 133 by multiplying the CVR of the distribution candidate advertisement content.

  Next, the extraction unit 135 extracts advertisement content to be distributed (step S107). Specifically, the extraction unit 135 calculates the CPC by multiplying the predicted CVR that is the corrected CVR corrected by the correction unit 134 and the target CPA. Then, the extracting unit 135 extracts advertisement content to be distributed based on the eCPM obtained by multiplying the calculated CPC and CTR. For example, the extraction unit 135 extracts, from among the distribution candidate advertisement contents, the one having the highest eCPM as the distribution target advertisement content.

  Finally, the distribution unit 136 distributes the advertisement content to be distributed extracted by the extraction unit 135 to the terminal device 10 (step S108).

[5. (Modification)
The information processing apparatus 100 according to the above embodiment may be implemented in various different forms other than the above embodiment. Therefore, in the following, another embodiment of the information processing apparatus 100 will be described.

[5-1. Correction considering category (1)]
In the above-described embodiment, an example in which the information processing apparatus 100 corrects the CVR by multiplying the CVR of the advertisement content by a weight value corresponding to the elapsed time has been described. However, the correction unit 134 of the information processing apparatus 100 may correct the CVR of the advertisement content using a weight value corresponding to the elapsed time in the category to which the distribution candidate advertisement content belongs.

  Here, FIG. 9 is a diagram illustrating the transition of the conversion rate according to the passage of time for each category to which the advertising content belongs. More specifically, FIG. 9 shows the conversion rate calculated from the number of times the advertising content distributed to the terminal device 10 is clicked by retargeting and the number of conversions converted from the clicked advertising content. It is plotted against the elapsed time divided into categories to which it belongs.

  In FIG. 9, FIG. 9A shows the transition of the conversion rate according to the passage of time of the advertising content belonging to the category “fashion”. FIG. 9B shows the transition of the conversion rate according to the elapsed time of the advertising content belonging to the category “cosmetics”. FIG. 9C shows transition of the conversion rate according to the passage of time of the advertising content belonging to the category “automobile”. FIG. 9D shows the transition of the conversion rate according to the elapsed time of the advertising content belonging to the category “reform”.

  Each category shown in FIGS. 9A to 9D is an example, and other categories may be set in the category to which the advertisement content belongs.

  Here, since the category of the transition of the conversion rate shown in FIG. 2 is not considered, it can be generally said that the conversion rate tends to decrease according to the elapsed time. For example, as shown in FIGS. 9A and 9B, in the advertising content belonging to the category “fashion” and the category “cosmetics”, the conversion rate decreases according to the elapsed time. However, as shown in FIG. 9C and FIG. 9D, in the advertising content belonging to the category “automobile” and the category “reform”, the conversion rate increases according to the elapsed time.

  For this reason, the conversion rate does not necessarily decrease with the elapsed time, and in the advertising content belonging to a certain category, the conversion rate may tend to increase with the elapsed time. . Therefore, the information processing apparatus 100 corrects the CVR of the advertisement content using a weight value corresponding to the elapsed time in the category to which the distribution candidate advertisement content belongs.

  Specifically, in the weight information storage unit 123-1 shown in FIG. 7, as described above, the weight value is set according to the transition of the conversion rate shown in FIG. A weight value is set according to the conversion transition for each category shown in FIG. Here, FIG. 10 is a diagram illustrating an example of the weight information storage unit 123-2 according to the modification.

  10A, FIG. 10A is a weight value set based on FIG. 9A showing the conversion rate transition of the advertising content belonging to the category “fashion”. In FIG. 9A, since the conversion rate is higher as the elapsed time is shorter, a weight value for increasing the degree of correction as the elapsed time is shorter, and a weight value for increasing CVR is set.

  FIG. 10B is a weight value set based on FIG. 9B showing the conversion rate transition of the advertising content belonging to the category “cosmetics”. In FIG. 9B, since the conversion rate is higher as the elapsed time is shorter, a weight value for increasing the degree of correction as the elapsed time is shorter, and a weight value for increasing CVR is set.

  FIG. 10C shows the weight values set based on FIG. 9C showing the conversion rate transition of the advertising content belonging to the category “automobile”. In FIG. 9C, since the conversion rate is increased according to the elapsed time, a weight value for increasing the degree of correction as the elapsed time is increased, and a weight value for increasing the CVR is set.

  FIG. 10D is a weight value set based on FIG. 9D showing the conversion rate transition of the advertising content belonging to the category “reform”. In FIG. 9D, since the conversion rate is increased according to the elapsed time, a weight value for increasing the degree of correction is set as the elapsed time is longer, and a weight value for increasing the CVR is set.

  For example, the straight line models M1a to M1d shown in FIGS. 9A to 9D are approximate curves calculated based on the variance of the plots in the respective drawings. The weight value is set based on

  In addition, when correction according to the elapsed time in each category is performed, a category is assigned in advance to each advertisement content stored in the advertisement content storage unit 122-1. Here, FIG. 11 is a diagram illustrating an example of the advertisement content storage unit 122-2 according to the modification. The advertisement content storage unit 122-2 illustrated in FIG. 11 corresponds to the advertisement content storage unit 122-1 illustrated in FIG. 6 further including an item “category”. Further, in order to simplify the description, each numerical value shown in the advertisement content storage unit 122-2 is the same as that of the advertisement content storage unit 122-1.

  “Category” indicates the category to which the corresponding advertisement content belongs as described above. The “category” is set, for example, when an advertisement content is submitted by an advertiser. The “category” may be arbitrarily set by the advertiser, or may be selected by the advertiser by the information processing apparatus 100 presenting candidates for “category”.

  In addition, the “category” may not be set for the advertiser, but may be automatically set by the information processing apparatus 100 for the submitted advertisement content. In such a case, the information processing apparatus 100 determines a category to be set, for example, by analyzing text, images, and the like corresponding to the advertising content. Below, an example of the correction process in this modification is demonstrated.

  Here, as in the above example, it is assumed that the receiving unit 132 receives an advertisement content acquisition request transmitted from the terminal device 10 of the user U1. As a result, the acquisition unit 133 specifies the advertisement contents AD11 and AD22 as distribution candidate advertisement contents, and acquires the elapsed time “53 hours” for the advertisement content AD11 and the elapsed time “11 hours” for the advertisement content AD22. And

  The correction unit 134 corrects the CVR of the advertisement content AD11 and AD22 using the weight value according to the elapsed time acquired by the acquisition unit 133 and the weight value in the category to which the advertisement content AD11 and AD22 belong respectively. . First, the correction unit 134 refers to the advertisement content storage unit 122-2 and identifies the category of each advertisement content. In the example of the advertising content storage unit 122-2 illustrated in FIG. 11, the correction unit 134 specifies the category “cosmetics” of the advertising content AD11 and the category “automobile” of the advertising content AD22.

  Next, the correcting unit 134 refers to the weight information storage unit 123-2 illustrated in FIG. 10 that corresponds to the identified category. Then, the correction unit 134 corrects the CVR of each of the advertising content AD11 and AD22 using the weight value corresponding to each of the elapsed times acquired by the acquisition unit 133.

  For the advertisement content AD11, the correction unit 134 refers to the weight information storage unit 123-2 illustrated in FIG. 10A and specifies the weight value “1.2” for the elapsed time “53 hours” corresponding to the advertisement content AD11. To do. Then, the correction unit 133 corrects the CVR of the advertisement content AD11 to “0.06” by multiplying the CVR “0.05” by the weight value “1.2”.

  For the advertisement content AD22, the correction unit 134 refers to the weight information storage unit 123-2 illustrated in FIG. 10C, and the weight value “1.3” for the elapsed time “11 hours” corresponding to the advertisement content AD22. Is identified. Then, the correction unit 133 corrects the CVR of the advertising content AD22 to “0.03” by multiplying the CVR “0.02” by the weight value “1.3”.

  In this way, the information processing apparatus 100 corrects the CVR that is the evaluation value of the advertisement content using the weight value according to the elapsed time in the category to which the distribution candidate advertisement content belongs. As shown in FIG. 9, the conversion rate has a tendency to increase or decrease according to the elapsed time depending on the category to which the advertising content belongs. In response to such a situation, the information processing apparatus 100 can correct the CVR by using the weight value corresponding to the conversion transition that is different for each category, thereby increasing the accuracy of the predicted CVR obtained by the correction. Can do.

[5-2. Correction considering category (2)]
The correction unit 134 may perform correction based on a model for each category to which the distribution candidate content belongs, and a value indicating the accuracy of the model based on the record information of the content belonging to the category. For example, the correction unit 134 sets distribution candidate contents belonging to a category whose value indicating the accuracy of the model is greater than or equal to a predetermined threshold as correction targets.

  As described with reference to FIG. 9, the plots for each category are obtained from the past performance information. The linear models M1a to M1d shown in FIGS. 9A to 9D show approximate curves calculated based on the variance of each plot. The numerical value in the upper right part of the graph is a determination coefficient indicating the accuracy of the approximate curve.

  Then, the closer the determination coefficient is to “1”, the higher the accuracy of the model, indicating that the change in the conversion rate according to the elapsed time can be explained with high accuracy. In other words, the lower the determination coefficient with respect to “1”, the lower the accuracy of the model and the lower the reliability of the conversion rate trend.

  For example, the determination coefficients of the models shown in FIG. 9A and FIG. 9C are “0.4” and “0.5”, and it can be said that the accuracy of the model is standard. The determination coefficient of the model shown in FIG. 9B is “0.7”, and it can be said that the accuracy of the model is relatively high. The determination coefficient of the model shown in FIG. 9D is “0.2”, and the accuracy of the model is low.

  For example, when a weight value is set based on a model with such low accuracy, the predicted CVR obtained by correction using the weight value is likely to have low accuracy (lack of reliability). For this reason, the correction unit 134 sets distribution candidate contents belonging to a category whose determination coefficient, which is a value indicating the accuracy of the model, equal to or higher than a predetermined threshold value, and belongs to a category whose determination coefficient is lower than the predetermined threshold value. Do not correct the content of the distribution candidate.

  In addition, since the correction unit 134 determines whether or not to correct, the information processing apparatus 100 further includes a model information storage unit 124 in this modification. As illustrated in FIG. 9, the model information storage unit 124 stores information on a model (approximate curve) calculated from the distribution of plots based on the distribution results. Here, FIG. 12 shows an example of the model information storage unit 124 according to the modification. In the example of FIG. 12, the model information storage unit 124 has items such as “category”, “model”, “determination coefficient”, and “correction flag”.

  “Category” indicates the category to which the advertising content belongs. The “model” indicates a model calculated from the distribution of plots based on the distribution result of the advertising content belonging to the corresponding “category”. Although a conceptual symbol is used here as a model, a model indicating a straight line such as “Y = αX + β” is actually stored. The “determination coefficient” indicates the accuracy of the model. The smaller the variance of the plot, the closer to “1”, and the larger the variance of the plot, the smaller the value from “1”.

  The “correction flag” indicates a flag for instructing whether or not to perform correction processing. In this modification, it is assumed that “0.4” is set as the threshold for the determination coefficient. In such a case, a correction flag “1” instructing to perform correction processing is associated with an advertising content belonging to a category having a determination coefficient equal to or greater than the threshold “0.4”. Further, a correction flag “0” instructing not to perform correction processing is associated with an advertisement content belonging to a category having a determination coefficient smaller than the threshold “0.4”.

  That is, FIG. 12 shows an example in which the model “M1a” and the determination coefficient “0.4” are calculated from the distribution results of the advertising content belonging to the category “fashion”. In addition, an example is shown in which an instruction is given to correct the CVR of the candidate advertisement content belonging to the category “fashion”.

  Note that the model and the determination coefficient may be calculated by an administrator who manages the information processing apparatus 100 by collecting performance information and stored in the model information storage unit 124, or the information processing apparatus 100 ( For example, it may be automatically calculated from the result information by a model generation unit). In addition, the threshold setting for the determination coefficient may be performed by an administrator who manages the information processing apparatus 100, for example. Below, an example of the correction process in this modification is demonstrated.

  Here, as in the above example, it is assumed that the receiving unit 132 receives an advertisement content acquisition request transmitted from the terminal device 10 of the user U1. As a result, the acquisition unit 133 specifies the advertisement contents AD11 and AD41 as distribution candidate advertisement contents, and acquires the elapsed time “53 hours” for the advertisement content AD11 and the elapsed time “20 hours” for the advertisement content AD41. And

  Here, the correction unit 134 determines whether or not the advertisement contents AD11 and AD41 are to be corrected. In determining, first, the correction unit 134 refers to the advertisement content storage unit 122-2 to identify the category of each advertisement content. In the example of the advertising content storage unit 122-2 illustrated in FIG. 11, the correction unit 134 specifies the category “cosmetics” of the advertising content AD11 and the category “reform” of the advertising content AD41.

  Next, the correction unit 134 refers to the model information storage unit 124 and determines whether or not the advertisement contents AD11 and AD41 are to be corrected. In the example of the model information storage unit 124 illustrated in FIG. 12, the correction flag “1” is associated with the category “cosmetics”. For this reason, the correction unit 134 determines that the advertising content AD11 belonging to the category “cosmetics” is to be corrected. Further, the correction flag “0” is associated with the category “reform”. For this reason, the correction unit 134 determines that the advertising content AD41 belonging to the category “reform” is not to be corrected.

  Since the correction unit 134 determines as described above, the correction unit 134 uses the weight value according to the elapsed time acquired by the acquisition unit 133 and the weight value in the category to which the advertising content AD11 belongs, to determine the advertising content AD11. Correct CVR. Since the subsequent correction processing is as described above, the description thereof will be omitted.

  As described above, the information processing apparatus 100 performs correction based on a value that indicates the accuracy of the model based on the record information of the content that belongs to the category to which the distribution candidate content belongs. Thereby, the information processing apparatus 100 can prevent the CVR from being corrected to a value with lower reliability.

[5-3. Elapsed time acquisition (1)]
In the above embodiment, an example has been described in which the acquisition unit 133 of the information processing apparatus 100 acquires the elapsed time since the user accessed the web page specified by the advertiser. However, the present invention is not limited to this example. For example, the acquisition unit 133 acquires the elapsed time since the user accessed the transition destination content that is transitioned when the advertising content is selected (for example, clicked or tapped). Also good. This point will be described with reference to FIG. Such transition destination content is generally called a landing page or the like.

FIG. 13 is a diagram illustrating an example of information processing according to the modification. As shown in FIG. 13, the landing page LP111 corresponds to a transition destination content that is transitioned when the advertisement content AD11 is clicked. In addition, the landing page LP111, Ritagetagu Tg15 ₁ is set. Further, it is assumed that the advertising content AD11 and the landing page LP111 are content managed by the advertiser C1.

The landing page LP211 corresponds to the transition destination content that is transitioned when the advertisement content AD21 is clicked. In addition, the landing page LP211, Ritagetagu Tg16 ₁ is set. Further, it is assumed that the advertisement content AD11 and the landing page LP211 are content managed by the advertiser C2.

Here, it is assumed that the terminal device 10 makes a transition from the advertising content AD11 to the landing page LP111 at “19:00 on March 13, 2016” in accordance with the operation of the user U2. That is, it is assumed that the terminal device 10 accesses the landing page LP111 at “19:00 on March 13, 2016” in accordance with the operation of the user U2 (step S21). The landing page LP111, from the fact that Ritagetagu Tg15 ₁ has been set, the terminal device 10, a user ID "U2", advertising ID "AD11", access date and time "at 2016 March 13, 19", an information processing apparatus 100 (step S22).

Further, it is assumed that the terminal device 10 makes a transition from the advertising content AD21 to the landing page LP211 at “14:00 on March 15, 2016” in accordance with the operation of the user U2. That is, it is assumed that the terminal device 10 accesses the landing page LP211 at “14:00 on March 15, 2016” in accordance with the operation of the user U2 (step S23). The landing page LP211, since it is set Ritagetagu Tg16 ₁ is, the terminal device 10, a user ID "U2", advertising ID "AD21", access date and time "at 2016 March 15, 14", an information processing apparatus It transmits to 100 (step S24).

  Here, it is assumed that the terminal device 10 transmits an advertisement content acquisition request based on accessing the landing pages P111 and P211 to the information processing device 100 according to the operation of the user U2 (step S25). Thereby, for example, if the acquisition unit 100 acquires the request reception date “March 15, 2016 22:00”, each elapsed time is used as a difference time between the request reception date and each access date and time. Is acquired (step S26). Further, the correction unit 134 corrects the CVR of the advertisement content AD11 corresponding to the landing page P111 and the CVR of the advertisement content AD21 corresponding to the landing page P211 according to the elapsed time acquired by the acquisition unit 133.

  Here, when the correction unit 134 corrects the CVR of the advertising content AD11 and the CVR of the advertising content AD21, either the advertising content AD11 or AD21 is again delivered to the user U2. That is, since the same advertisement may be continuously delivered to the user, the impression by the user may be deteriorated.

  For this reason, for example, when the advertiser C1 submits an advertising content other than the advertising content AD11, the correcting unit 134 may set the advertising content other than the advertising content AD11 as a correction target. Similarly, for example, when the advertiser C2 has submitted an advertising content other than the advertising content AD21, the correcting unit 134 may set the advertising content other than the advertising content AD21 as a correction target.

[5-4. Elapsed time acquisition (2)]
Further, the acquisition unit 133 may acquire an elapsed time since the user performed an action such as requesting a material or purchasing a product on a page where the product is purchased (hereinafter referred to as “purchase page”). This point will be described with reference to FIG.

FIG. 14 is a diagram illustrating an example of information processing according to the modification. As illustrated in FIG. 14, the purchase completion page EC111 is a page that is transitioned to when the purchase of the product is determined on the purchase page EC11. This is a purchase completion page that includes information about. In addition, the purchase completion page EC111, Ritagetagu Tg15 ₂ is set. The purchase page EC11 and the purchase completion page EC111 are contents managed by the advertiser C1.

The purchase completion page EC211 is a page that is transitioned when purchase of a product is determined on the purchase page EC21, and includes a purchase completion page that includes information about the purchased product such as a product name, purchase price, settlement method, and delivery information. It is. In addition, the purchase completion page EC211, Ritagetagu Tg16 ₂ is set. Further, it is assumed that the purchase page EC21 and the purchase completion page EC211 are contents managed by the advertiser C2.

Here, it is assumed that the terminal device 10 transits from the purchase page EC11 to the purchase completion page EC111 at “19:00 on March 13, 2016” in accordance with the operation of the user U2. That is, it is assumed that the terminal device 10 accesses the product completion page EC111 at “19:00 on March 13, 2016” by purchasing the product by the user U2 pressing the purchase button B1 (step S31). ). The commodity complete page EC111, Ritagetagu Tg15 from the fact _{that 2} is set, the terminal device 10, a user ID "U2", advertiser ID "C1", access date and time "at 2016 March 13, 19" information It transmits to the processing apparatus 100 (step S32).

Further, it is assumed that the terminal device 10 makes a transition from the purchase page EC21 to the purchase completion page EC211 at “14:00 on March 15, 2016” in accordance with the operation of the user U2. That is, it is assumed that the terminal device 10 accesses the product completion page EC211 at “14:00 on March 15, 2016” by purchasing the product by pressing the purchase button B2 by the user U2 (step S33). ). The commodity complete page EC211, Ritagetagu Tg16 from the fact _{that 2} is set, the terminal device 10, a user ID "U2", advertiser ID "C2", access date and time "at 2016 March 15, 14" information It transmits to the processing apparatus 100 (step S34).

  Here, it is assumed that the terminal device 10 transmits an advertisement content acquisition request based on the access to the product completion pages EC111 and EC211 to the information processing device 100 according to the operation of the user U2 (step S35). Thereby, for example, if the acquisition unit 100 acquires the request reception date “March 15, 2016 22:00”, each elapsed time is used as a difference time between the request reception date and each access date and time. Is acquired (step S36).

Further, the correction unit 134 corrects the CVR of the advertising content submitted by the advertisers C1 and C2 according to the elapsed time acquired by the acquisition unit 133. In the example of the advertisement content storage unit 122-1 illustrated in FIG. 6, the advertisement contents AD11 and AD21 are submitted to the advertiser C1, and the advertisement contents AD21 and AD22 are submitted to the advertiser C2. In such an example, the correction unit 134 corrects the CVR of each of the four advertisement contents.
[5-5. (Evaluation value)
So far, the correction unit 134 of the information processing apparatus 100 has been described as an example of correcting the CVR of the advertising content as the evaluation value. However, the correction unit 134 may correct evaluation values other than CVR. For example, the correction unit 134 may correct CTR and eCPM.

  For example, since the information processing apparatus 100 has a conversion optimization function, an example of correcting CVR to automatically calculate CTR on behalf of an advertiser has been described. However, the information processing apparatus 100 may perform a process of determining the advertisement content to be distributed based on the bid price and the eCPM calculated using the CTR by causing the advertiser to set the bid price.

  In such a case, the information processing apparatus 100 may predict CTR. If the accuracy of the predicted CTR predicted is low, the calculated eCPM is likely to be low in accuracy. Therefore, the information processing apparatus 100 applies the correction process described above to eCPM.

[5-6. About extraction process)
In the above-described embodiment, an example in which the extraction unit 135 of the information processing apparatus 100 calculates eCPM using the predicted CVR obtained by the correction by the correction unit 134 and distributes the advertising content having the highest calculated eCPM is shown. However, without being limited to this example, the extraction unit 135 may periodically extract the advertising content having the highest predicted CVR as a distribution target. A high predicted CVR means that conversion is more likely to be obtained. For this reason, the information processing apparatus 100 can improve the advertiser satisfaction by distributing the advertising content with the highest predicted CVR as well as the eCPM.

[6. program〕
Further, the information processing apparatus 100 according to the embodiment described above is realized by a computer 1000 having a configuration as shown in FIG. 13, for example. FIG. 13 is a hardware configuration diagram illustrating an example of a computer 1000 that implements the functions of the information processing apparatus 100. The computer 1000 includes a CPU 1100, RAM 1200, ROM 1300, HDD 1400, communication interface (I / F) 1500, input / output interface (I / F) 1600, and media interface (I / F) 1700.

  The CPU 1100 operates based on a program stored in the ROM 1300 or the HDD 1400 and controls each unit. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started up, a program depending on the hardware of the computer 1000, and the like.

  The HDD 1400 stores programs executed by the CPU 1100, data used by the programs, and the like. The communication interface 1500 receives data from other devices via the communication network 50 and sends the data to the CPU 1100, and transmits the data generated by the CPU 1100 to other devices via the communication network 50.

  The CPU 1100 controls an output device such as a display and a printer and an input device such as a keyboard and a mouse via the input / output interface 1600. The CPU 1100 acquires data from the input device via the input / output interface 1600. Further, the CPU 1100 outputs the generated data to the output device via the input / output interface 1600.

  The media interface 1700 reads a program or data stored in the recording medium 1800 and provides it to the CPU 1100 via the RAM 1200. The CPU 1100 loads such a program from the recording medium 1800 onto the RAM 1200 via the media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or PD (Phase change rewritable disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. Etc.

  For example, when the computer 1000 functions as the information processing apparatus 100 according to the embodiment, the CPU 1100 of the computer 1000 implements the function of the control unit 130 by executing a program loaded on the RAM 1200. The HDD 1400 stores data in the storage unit 120. The CPU 1100 of the computer 1000 reads these programs from the recording medium 1800 and executes them, but as another example, these programs may be acquired from other devices via the communication network 50.

[7. Others]
Of the processes described in the above embodiment, all or part of the processes described as being automatically performed can be performed manually, or all of the processes described as being performed manually or A part can be automatically performed by a known method. In addition, the processing procedures, specific names, and information including various data and parameters shown in the document and drawings can be arbitrarily changed unless otherwise specified.

  Further, each component of each illustrated apparatus is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution / integration of each device is not limited to that shown in the figure, and all or a part thereof may be functionally or physically distributed or arbitrarily distributed in arbitrary units according to various loads or usage conditions. Can be integrated and configured.

  Moreover, each embodiment mentioned above can be combined suitably in the range which does not contradict a process content.

[8. effect〕
The information processing apparatus 100 according to the embodiment includes an acquisition unit 133 and a correction unit 134. The acquisition unit 133 acquires the elapsed time since the user took action related to the predetermined content. The correction unit 134 corrects the evaluation value of the content of the distribution candidate based on the action according to the elapsed time acquired by the acquisition unit 133.

  As described above, the information processing apparatus 100 according to the embodiment can apply the correction according to the elapsed time to the evaluation value that is simply calculated from the record information, and thus can improve the accuracy of the evaluation value of the content.

  In the information processing apparatus 100 according to the embodiment, the acquisition unit 133 acquires the elapsed time since the user accessed predetermined content as the elapsed time, and the correction unit 134 is the content of the distribution candidate based on the action. As described above, the evaluation value of the advertisement content corresponding to the predetermined content accessed by the user is corrected.

  As described above, the information processing apparatus 100 according to the embodiment can apply a correction according to the elapsed time since the user accessed the predetermined content to the evaluation value simply calculated from the performance information. The accuracy of the evaluation value can be increased.

  In the information processing apparatus 100 according to the embodiment, the reception unit 132 receives a content acquisition request based on a user accessing a predetermined content from the user's terminal device, and the acquisition unit 133 receives the reception unit 132. The elapsed time elapsed until the present time when the acquisition request is received is acquired.

  As described above, the information processing apparatus 100 according to the embodiment responds to the evaluation value simply calculated from the performance information according to the elapsed time from when the user accesses the predetermined content until the advertisement content acquisition request is received. Therefore, the accuracy of the content evaluation value can be increased.

  In the information processing apparatus 100 according to the embodiment, the correction unit 134 corrects the content evaluation value using a predetermined weight value corresponding to the elapsed time.

  Thereby, the information processing apparatus 100 according to the embodiment can increase the accuracy of the evaluation value of the content.

  In the information processing apparatus 100 according to the embodiment, the correction unit 134 corrects the content evaluation value using a higher weight value as the elapsed time is shorter.

  In general, the evaluation value (for example, the conversion rate) tends to be higher as the elapsed time is shorter. For this reason, in the information processing apparatus 100, since correction according to such a tendency can be applied, the accuracy of the evaluation value of the content can be increased.

  In the information processing apparatus 100 according to the embodiment, the correction unit 134 corrects the evaluation value of the content using a predetermined weight value corresponding to the elapsed time in the category to which the distribution candidate content belongs.

  In general, evaluation values (for example, conversion rate) are short in elapsed time and show a tendency to decrease according to elapsed time. However, if evaluation values according to elapsed time are aggregated into categories, Some categories show a tendency that the evaluation value increases. Since the information processing apparatus 100 can perform correction according to the tendency for each category in this way, the accuracy of the content evaluation value can be further increased.

  In the information processing apparatus 100 according to the embodiment, the correction unit 134 is a model for each category to which the content of the distribution candidate belongs, and a value indicating the accuracy of the model based on the performance information of the content belonging to the category is a predetermined value. Content of distribution candidates belonging to a category equal to or higher than the threshold is set as a correction target.

  For example, when correction is performed using a weight value based on a model with low accuracy, there is a high possibility that the accuracy of the evaluation value obtained by the correction is low. Therefore, the information processing apparatus 100 can prevent an evaluation value with low accuracy from being calculated by setting only a content whose value indicating the accuracy of the model belongs to a category that is equal to or greater than a predetermined threshold. .

  As described above, some of the embodiments of the present application have been described in detail based on the drawings. It is possible to implement the present invention in other forms with improvements.

  In addition, the “section (module, unit)” described above can be read as “means” or “circuit”. For example, the receiving unit can be read as receiving means or a receiving circuit.

DESCRIPTION OF SYMBOLS 1 Information processing system 10 Terminal apparatus 20 Advertiser apparatus 30 Information provision apparatus 100 Information processing apparatus 120 Storage part 121 Time information storage part 122-1 Advertisement content storage part 123-1 Weight information storage part 130 Control part 132 Reception part 133 Acquisition part 134 Correction Unit 135 Extraction Unit 136 Distribution Unit

Claims (10)

An acquisition unit for acquiring an elapsed time since the user took an action related to predetermined content;
A correction device comprising: a correction unit that corrects the evaluation value of the content of the distribution candidate based on the behavior according to the elapsed time acquired by the acquisition unit. The acquisition unit acquires the elapsed time since the user accessed the predetermined content as the elapsed time,
The correction device according to claim 1, wherein the correction unit corrects an evaluation value of an advertisement content corresponding to a predetermined content accessed by the user as content of a delivery candidate based on the action. A receiving unit for receiving a content acquisition request based on the user accessing the predetermined content from the user terminal device;
The correction apparatus according to claim 1, wherein the acquisition unit acquires the elapsed time that has elapsed until the current time when the acquisition request is received by the reception unit. The correction device according to claim 1, wherein the correction unit corrects the evaluation value by using a predetermined weight value corresponding to the elapsed time. The correction device according to claim 4, wherein the correction unit corrects the evaluation value using a higher weight value as the elapsed time is shorter. The correction device according to claim 4, wherein the correction unit corrects the evaluation value using a predetermined weight value corresponding to the elapsed time in a category to which the content of the distribution candidate belongs. The correction unit is a model for each category to which the content of the distribution candidate belongs, and a value indicating the accuracy of the model based on the record information of the content belonging to the category is a value of the distribution candidate belonging to the category having a predetermined threshold value or more. The correction device according to any one of claims 4 to 6, wherein the content is a correction target. The said correction | amendment part correct | amends at least any one among the conversion rate, click rate, or profit expectation value of advertisement content of a delivery candidate as an evaluation value of the said delivery candidate content. The correction apparatus as described in any one of these. A correction method executed by a correction device,
An acquisition step of acquiring the elapsed time since the user took action on the predetermined content;
And a correction step of correcting the evaluation value of the content of the delivery candidate based on the behavior according to the elapsed time acquired by the acquisition step. An acquisition procedure for acquiring the elapsed time since the user took action on the predetermined content;
A correction program for causing a computer to execute a correction procedure for correcting an evaluation value of content of a delivery candidate based on the behavior according to an elapsed time acquired by the acquisition procedure.

Images