JP6635457B1 - Information processing apparatus, program, and system - Google Patents

Abstract

An information processing apparatus, system, and program capable of efficiently verifying the effect of an advertisement without paying attention to statistics are provided. In a system in which a user terminal and an information processing device are connected via a network, a storage unit of the information processing device stores element data in association with a module. The guide information generating unit 331 is configured such that the total number of combinations of the element data for each module, the number of verifications verified by the orthogonal array table, and the plurality of selectable element data linked to the module are one by one. Generate a list of tables. The communication unit 31 transmits the guide information to the user terminal and receives the orthogonal array table selected by the user. The advertisement generation unit 333 generates an advertisement candidate having the verification number specified in the orthogonal array table. [Selection diagram] FIG.

Description

  The present invention relates to an information processing device, a program, and a system.

  2. Description of the Related Art Conventionally, as an advertisement proposal system, for a display content of a web page, a user's preference is analyzed by a so-called A / B test, and a target is set for each of a plurality of modules (buttons, characters, etc.) configuring the display content. 2. Description of the Related Art An A / B test measurement management system that automatically switches a plurality of modules to achieve a target is known (for example, Patent Document 1).

JP-A-2005-225361 JP 2008-502079 A

  However, the above-described conventional A / B test measurement management system has a configuration in which modules are changed in order to achieve a target set for each module, but describes how to select a module for verification. There was no. It is also believed that making such a selection artificially requires enormous amounts of time and effort. In addition, devising an efficient method requires knowledge of statistics, etc., but it is relatively burdensome to understand statistics compared to the need to verify the effectiveness of advertising related to your business. It can be said that it is heavy.

  The present invention has been made in view of such circumstances, and it is an object of the present invention to provide an information processing apparatus, a system, and a program capable of efficiently verifying the effect of an advertisement without being conscious of statistics.

  According to the present invention, there is provided an information processing apparatus for generating a plurality of advertisement candidates which are candidates for a desired advertisement, wherein data constituting the advertisement is defined as element data, and When a module classified and summarized based on a rule defined by at least one of a name and an idea is defined as a module, the information processing apparatus includes a storage unit, a guide information generation unit, a communication unit, and an advertisement generation unit. The storage unit stores the element data in association with the module, and the guide information generation unit is configured to be able to generate at least one guide information, wherein the guide information is the module The total number of combinations of the element data for each, the number of verifications verified by the orthogonal array table, and one or more selectable element data linked to the module The communication unit is configured to be able to transmit the guide information to a user terminal of the user, configured to be able to receive the orthogonal array selected by the user from the user terminal, An information processing apparatus is provided, wherein the generation unit is configured to be able to generate an advertisement candidate having a verification number specified in the orthogonal array table.

  According to the information processing apparatus according to the present invention, the guide information generation unit generates at least one guide information. In the guide information, the total number of combinations of element data for each module, the number of verifications verified by the orthogonal array table, and a plurality of selectable element data linked to the module are listed one by one. Table and included. Then, by transmitting this to the user terminal by the communication unit, the advantageous effect that the user can appropriately verify the effect of the advertisement based on the guide information even if the user does not have knowledge of statistics is provided. Play.

FIG. 1 is a functional block diagram illustrating an outline of a system according to an embodiment. FIG. 2 is a functional block diagram illustrating an outline of a user terminal. FIG. 2 is a functional block diagram illustrating an outline of an information processing apparatus. An example of a module constituting an advertisement and element data included in the module. An example of a module constituting an advertisement and element data included in the module. An example of guide information including information on an orthogonal array that is easy for a user to understand. 9 is a flowchart illustrating a method for verifying the effect of an advertisement. One example of a conventional orthogonal array table (cited from Patent Document 2). The figure which shows the aspect which can produce | generate element data with one button. An example of a verification result.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. Various features shown in the embodiments described below can be combined with each other. In particular, in this specification, the term “unit” may include, for example, a combination of hardware resources implemented by circuits in a broad sense and information processing of software that can be specifically realized by these hardware resources. . In the present embodiment, various types of information are handled. These pieces of information are represented by a signal value as a binary bit set composed of 0 or 1, and communication / operation is executed on a circuit in a broad sense. Can be done.

  The circuit in a broad sense is a circuit realized by at least appropriately combining a circuit (Circuit), circuits (Circuitry), a processor (Processor), a memory (Memory), and the like. That is, an application-specific integrated circuit (ASIC), a programmable logic device (for example, a Simple Programmable Logic Device: SPLD), a composite programmable logic device (Complex Programable DCL, and a Digital Pledge DCL). It includes a programmable gate array (Field Programmable Gate Array: FPGA) and the like.

  In this specification, an advertisement includes element data E, an Internet advertisement, a television advertisement, a paper medium advertisement such as a direct mail or an insertion flyer printed from text data, and an e-mail composed of text data. Includes e-mail magazine ads. Specifically, the element data E can include text data, image data, vector data, effect information, or audio data.

  In this specification, text data refers to data that can be identified when a text search is performed on a main body displayed on a display screen or the like. The image data is data of a format in which a main body displayed on a display screen or the like is a group of pixels and does not hit a text search. Further, the effect information is information on effects of text data, image data, or vector data on the advertisement, and includes a movement start position, a start expression effect, a movement end position, an end expression effect, and the like. The expression effects include effects such as slide-in, dissolve-in, deformation, flashing, and blurring. Of course, it should be noted that the present invention is not limited to this.

1. Overall Configuration In the first section, the overall configuration of the system 1 according to the present embodiment will be described. FIG. 1 is a functional block diagram illustrating an outline of a system 1 according to the embodiment. The system 1 is a system including a user terminal 2 and an information processing device 3, which are connected via a telecommunication line (network). Hereinafter, each component will be described in more detail.

1.1 User terminal 2
FIG. 2 is a functional block diagram illustrating a configuration outline of the user terminal 2. The user terminal 2 includes a communication unit 21, a storage unit 22, a control unit 23, a display unit 24, and an input unit 25, and these components are electrically connected via a communication bus 20. I have. Hereinafter, each component will be further described.

<Communication unit 21>
The communication unit 21 preferably uses a wired communication means such as USB, IEEE1394, Thunderbolt, and wired LAN network communication, but performs wireless LAN network communication, mobile communication such as LTE / 3G, Bluetooth (registered trademark) communication as necessary. May be included. That is, it is more preferable to implement as a set of these plurality of communication means. In particular, the original advertisement data OA, which is an advertisement composed of at least one of text data, image data, vector data, effect information, audio data, and the like, can be transmitted to the information processing device 3 described below. The advertisement verification result R from the information processing device 3 can be received.

<Storage unit 22>
The storage unit 22 stores various information defined by the above description. This is, for example, as a storage device such as a solid state drive (Solid State Drive: SSD) or a random access memory (Random Access Memory :) for storing temporarily necessary information (arguments, arrays, etc.) related to the operation of the program. RAM) or the like. Also, a combination of these may be used. In particular, the storage unit 22 stores various programs and the like related to the user terminal 2 executed by the control unit 23. For example, the storage unit 22 includes a graphical display for the user U to input original advertisement data OA that is at least one of text data, image data, vector data, effect information, audio data, or the like via the input unit 25 described below. -A program related to a user interface (GUI) is stored.

<Control unit 23>
The control unit 23 performs processing and control of the entire operation related to the user terminal 2. The control unit 23 is, for example, a central processing unit (CPU) (not shown). The control unit 23 realizes various functions related to the user terminal 2 by reading a predetermined program stored in the storage unit 22. That is, a specific function can be executed by the information processing by software (stored in the storage unit 22) being specifically realized by the hardware (control unit 23). In FIG. 2, the control unit 23 is described as a single control unit 23. However, the present invention is not limited to this, and a plurality of control units 23 may be provided for each function. Also, a combination thereof may be used.

<Display unit 24>
The display unit 24 may be included in the user terminal 2 itself, or may be externally provided close to the user terminal 2. It is preferable that display devices such as a CRT display, a liquid crystal display, an organic EL display, and a plasma display, which are internally / externally mounted, are used separately as needed. The display unit 24 can selectively display a GUI screen in response to a predetermined control signal from the control unit 23. To supplement the GUI, for example, a specific interface such as an icon, a slide bar, a radio button, or the like may be displayed.

<Input unit 25>
The input unit 25 may be included in, for example, the user terminal 2 itself, or may be externally provided close to the user terminal 2. In particular, the input unit 25 may appropriately employ a user interface such as a switch button, a mouse, a keyboard, or a voice input, or may be implemented as a touch panel provided on the display unit 24 in a superimposed manner. An instruction (command) of the user is received via the input unit 25. The instruction is transferred to the control unit 23 via the communication bus 20, and the control unit 23 can execute predetermined control and calculation as needed. Particularly, in the present embodiment, the input unit 25 is used when the user U inputs the original advertisement data OA.

1.2 Information processing device 3
FIG. 3 is a functional block diagram illustrating an outline of the configuration of the information processing apparatus 3. The information processing device 3 includes a communication unit 31, a storage unit 32, and a control unit 33, and these components are electrically connected inside the information processing device 3 via a communication bus 30. Hereinafter, each component will be further described.

<Communication unit 31>
The communication unit 31 preferably uses wired communication means such as USB, IEEE1394, Thunderbolt, and wired LAN network communication, but performs wireless LAN network communication, mobile communication such as LTE / 3G, Bluetooth (registered trademark) communication as necessary. May be included. That is, it is more preferable to implement as a set of these plurality of communication means. In particular, it is configured to be able to receive original advertisement data OA (a set of text data or image data), which is an advertisement, from the user terminal 2 and to be able to transmit a verification result R of the effect of the advertisement to the user terminal 2. Good. Furthermore, guide information G described below may be configured to be transmitted to the user terminal 2, and the user terminal 2 may be configured to be able to receive the orthogonal array table selected by the user U. Further, the communication unit 31 may be configured to be able to execute each function by the control unit 33 to be described later from the user terminal 2 via a predetermined GUI, and to be able to communicate with the user terminal 2.

<Storage unit 32>
The storage unit 32 stores various information defined by the above description. This is, for example, as a storage device such as a solid state drive (Solid State Drive: SSD) or a random access memory (Random Access Memory :) for storing temporarily necessary information (arguments, arrays, etc.) related to the operation of the program. RAM) or the like. Also, a combination of these may be used. The storage unit 32 stores the original advertisement data OA. Further, it should be noted that the storage unit 32 stores the element data E in association with the module M. The module M is obtained by classifying and summarizing the element data E based on a rule defined by at least one of appearance, name, and idea. This is described in more detail in Section 2.

  The storage unit 32 is an advertisement in which the advertisement analysis unit 330 in the control unit 33 performs a morphological analysis on the original advertisement data OA to extract the element data E and proposes new element data E corresponding to each element data E. The analysis program is stored. The storage unit 32 further stores a program that causes the guide information generation unit 331 in the control unit 33 to generate at least one guide information G. The storage unit 32 determines the combination of the element data E obtained by selecting the element data E from each module M by the combination determination unit 332 in the control unit 33, and sets the determined non-all combinations. The program for performing the ranking is further stored. The storage unit 32 further stores a program in which the advertisement generation unit 333 in the control unit 33 generates a plurality of advertisements defined by the above-mentioned non-all combinations in the order determined by the combination determination unit 332. I do. The storage unit 32 further stores a program such that the effect verification unit 334 of the control unit 33 performs the effect verification of the plurality of advertisements generated by the advertisement generation unit 333.

<Control unit 33>
The control unit 33 performs processing and control of the entire operation related to the information processing device 3. The control unit 33 is, for example, a central processing unit (CPU) (not shown). The control unit 33 realizes various functions related to the information processing device 3 by reading a predetermined program stored in the storage unit 32. That is, the information processing by software (stored in the storage unit 32) is specifically realized by the hardware (control unit 33), so that the advertisement analysis unit 330, the combination determination unit 332, the advertisement generation unit 333, and It can be executed as the effect verification unit 334. Hereinafter, the advertisement analysis unit 330, the combination determination unit 332, the advertisement generation unit 333, and the effect verification unit 334 will be described in further detail.

[Ad Analysis Unit 330]
In the advertisement analysis unit 330, information processing by software (stored in the storage unit 32) is specifically realized by hardware (control unit 33). The advertisement analysis unit 330 analyzes the original advertisement data OA received from the user terminal 2 via the communication unit 31. Specifically, the original advertisement data OA is decomposed into original advertisement text data OAT as text data and original advertisement image data OAI as image data. The original advertisement image data OAI is further decomposed into each element data E, for example, into each element data E of a background image BI, a logo mark LG, a main image MI, a sub image SI, a title image TL, and a description image EP. It is disassembled (see FIG. 4). Of course, titles and catch phrases may be included. Also, from the original advertisement text data OAT, element data E serving as a keyword is extracted by morphological analysis (see FIG. 5A). Further, for each element data E, new element data Ea is proposed based on a rule defined by at least one of the appearance, name, and idea of the element data E, and element data E, Ea based on the rule are proposed. Are defined as a module M. These are described in more detail in Section 2.

[Guide information generation unit 331]
The guide information generation unit 331 is one in which information processing by software (stored in the storage unit 32) is specifically realized by hardware (the control unit 33). The guide information generating unit 331 is configured to generate at least one guide information G. Here, the guide information G includes the total number of combinations of the element data E for each module M, the number of verifications verified by the orthogonal array table, and one or more selectable element data E linked to the module M. And a table T arranged one by one. These are described in more detail in Section 2.

[Combination determination unit 332]
The combination determining unit 332 is one in which information processing by software (stored in the storage unit 32) is specifically realized by hardware (the control unit 33). The combination determining unit 332 selects at least two modules M stored in the storage unit 32. Then, the combination is determined by selecting the element data E included in them. For example, assuming that there are three modules M and each of the modules M has three element data E, there are 27 combinations of 3 to the third power. However, it is not preferable to perform all types of verification in consideration of advertisement traffic and verification costs.

  The combination determining unit 332 determines, based on statistics, a combination to be actually verified among the candidate combinations. Specifically, the combination determining unit 332 determines a preferable orthogonal array table based on the number of element data E, and determines non-all combinations based on the determined orthogonal array table. This is described in more detail in Section 2. Then, the combination determination unit 332 ranks the non-all combinations in accordance with a predetermined rule. For example, it is preferable to rank the advertisements in the order in which the effectiveness of the advertisement is predicted to be high, based on the verification results already performed in the past. Further, it is preferable that the results of the verifications already performed in the past are machine-learned so that an effective advertisement can be predicted.

[Advertising Generator 333]
The advertisement generation unit 333 is one in which information processing by software (stored in the storage unit 32) is specifically realized by hardware (control unit 33). The advertisement generation unit 333 sequentially generates a plurality of advertisement candidates defined by the above-mentioned non-all combinations in the order determined by the combination determination unit 332. Here, instead of generating the advertisements (advertisements more accurately) related to the above-mentioned non-all combinations at the same time, only the advertisements (at least two) to be verified at the same time need be generated. A new advertisement may be generated as needed when performing the verification.

[Effect verification unit 334]
The effect verification unit 334 is one in which information processing by software (stored in the storage unit 32) is specifically realized by hardware (the control unit 33). The effect verification unit 334 verifies the effect of the plurality of advertisements generated by the advertisement generation unit 333. In particular, it is preferable to perform at least one of the following (1) and (2) for at least two advertisements to be verified simultaneously. See Section 2 for effect measures. It should also be noted that such an effect verification includes a verification that at least two advertisements have a statistically sufficient difference. In other words, the following threshold value is set so that it can be verified whether a statistically sufficient difference has been obtained.
(1) If there is an advertisement whose effect measure does not satisfy the threshold value, stop the advertisement. (2) If any advertisement whose effect measure does not satisfy the threshold value, stop all advertisements.

  The advertisement generation unit 333 and the effect verification unit 334 may be composed of a plurality of units, and may be configured to verify the effects of a plurality of advertisements defined by at least two of the non-all combinations in a parallel operation. Then, by repeating such processing, finally, what combination is effective (or not effective) is derived as the verification result R. The verification result R is transmitted from the communication unit 31 in the information processing device 3 to the user terminal 2 via the electric communication line, and the user U can confirm the verification result R on the display unit 24.

  More specifically, the effect verifying unit 334 is configured to perform a multivariate analysis on the number of advertisement candidates to be verified and generate a verification result R (effect data) indicating an effect for each element data E. The communication unit 31 is configured to be able to transmit the verification result R to the user terminal 2 of the user U. This allows the user U to check the verification result R as a report on the display unit 24 and perform analysis.

2. Detailed Description of Information In the second section, information processed in the system 1 described in the first section will be described in detail.

2.1 Module M and Element Data E
4 and 5A show an example of a module M that constitutes an advertisement and element data E included in the module M. As shown in FIG. 4, original advertisement image data OAI, which is a certain web advertisement, is obtained from element data E including a background image BI, a logo mark LG, a main image MI, a sub image SI, a title image TL, and a description image EP. Become. The element data E is obtained by disassembling the original advertisement image data OAI by the advertisement analysis unit 330 in the control unit 33. However, the user U may directly select and input the element data E by the user terminal 2. .

  FIGS. 5A and 5B show a case relating to the original advertisement text data OAT described below the original advertisement image data OAI. For example, a portion indicated by a bold underline in FIG. 5A is extracted as element data E based on the result of the morphological analysis. More specifically, the original advertisement text data OAT selected / input by the user U is subjected to morphological analysis by the advertisement analysis unit 330 in the control unit 33, and each element data E is extracted.

  Further, as shown in FIG. 5B, new element data Ea corresponding to each element data E can be proposed by the advertisement analysis unit 330. It is preferable that the new element data Ea proposed here is proposed based on a rule defined by at least one of the original element data E and appearance, name, and idea. For example, in FIG. 5B, element data Ea having a similar idea is proposed. Specifically, for the first element data E1 “job introduction”, the first element data E1a “job search” is proposed. For the second element data E2, “registration free”, the second element data E2a, “registration 0 yen” has been proposed. A third element data E3a, "10,000", has been proposed for the third element data E3, "10,000". For the fourth element data E4, “job offer”, the fourth element data E4a, “Shigoto” has been proposed. In this way, the first to fourth modules M1 to M4 as shown in FIG. 5B are defined, and the element data E is linked to the module M and stored in the storage unit 32 of the information processing device 3. . In other words, new element data Ea corresponding to each of the plurality of decomposed element data E is proposed, and the module M is defined by the original element data E and the newly proposed element data Ea.

2.2 Effect Measure The effect measure is defined as a function used in verifying the effect of the advertisement, and may have a plurality of parameters as variables. In the effect verification on the web, for example, the number of inflows by clicks, the number of inquiries by the advertisement, the unit price of induction, the unit price of acquisition, and the like can be variables. In other words, the effect of the advertisement is verified by performing the tallying regarding the effect measure in various combinations for a certain period of time.

  The target of such an effect measure may be a so-called closed member such as a user in a company or a predetermined small or large number of users, or an unspecified large number of so-called open members. Also, instead of simply looking at the number of clicks, etc., a questionnaire is conducted, and evaluation such as “good”, “bad”, “neither” or the like is performed. May be implemented so as to be able to accumulate.

2.3 Non-all combinations based on orthogonal array table For example, consider first to fourth modules M1 to M4 shown in FIG. 5B. Since each module M includes two element data E, there are 16 combinations of all the combinations 2 to the fourth power. However, it is not preferable to perform all types of verification in consideration of advertisement traffic and verification costs. In the present embodiment, as described above, the combination determining unit 332 determines a preferable orthogonal array table, and based on the determined orthogonal array table, determines non-all combinations in which effect verification is actually performed. As the module M selected here and the element data E included in the module M, those which have obtained good results in the effect verification performed in the past (the value of the effect measure was high in the past) are positively adopted. Good. Further, it is preferable that not only the results of the past verification be stored in the storage unit 32 but also those results are machine-learned to make a preferable selection of the module M and the element data E.

  There are various types of orthogonal array tables. In the system 1 according to the present embodiment, based on the number of modules M and the number of element data E included in the modules M, the combination determining unit 332 performs appropriate orthogonal allocation based on statistics. Preferably, a column table can be selected. Of course, the user U may be able to select a preferable orthogonal array table via the input unit 25 of the user terminal 2.

  Incidentally, the orthogonal array table is originally a table showing actual non-all combinations as shown in FIG. However, the user only needs to know the combination of the module M and the element data E that make up the finally preferred advertisement, rather than the combination of the module M and the element data E that make up each advertisement candidate. It is preferable that the degree of efficiency be presented in an easy-to-understand manner.

  Therefore, in the present embodiment, as shown in FIGS. 6A and 6B, guide information G may be displayed on the display unit 24 of the user terminal 2 in a manner that is easy for the user U to understand. The guide information G includes the total number of combinations of the element data E for each module M, the number of verifications verified by the orthogonal array table, and one of a plurality of selectable element data E linked to the module M. And tables to be listed. The user U can confirm such information in a state of having a list via the user terminal 2 and select a preferable orthogonal array table. In addition, the guide information G may include the ratio of the number of combinations of all combinations and the number of combinations reduced by the orthogonal array table in order for the user U to grasp the efficiency (reduction efficiency) of verification.

Further, additional information such as a warning may be displayed on the display unit 24 of the user terminal 2. For example, the information processing device 3 may include a warning unit (not shown) in which information processing by software (stored in the storage unit 32) is specifically realized by hardware (the control unit 33). For example, when the orthogonal array table selected by the user U is not suitable for the verification to be performed in the future, it is configured to be able to generate warning information according to at least one of the following (1) to (4).
(1) Insufficient number of modules M or element data E (2) Insufficient number of modules M or element data E (3) Candidate module M or element data E to be deleted (4) The module M or the candidate for the element data E to be added. Then, the warning information is transmitted to the user terminal 2 via the communication unit 31.

  More specifically, "Please delete one element data E for each of the modules M1 to M3", "Please add one element data E to each of the modules M1 to M3", " Without changing the number of tests, three pieces of element data can be added in two modules M, or six pieces of element data can be added in one module M. " Further, it is preferable that deletion, additional creation, and the like can be performed with a single button from a screen displaying such accompanying information. For example, FIG. 9 shows a mode in which the generation of the missing element data E can be performed with one button.

  In addition, it is preferable that a maximum verifiable number of advertisements is first estimated from the traffic, and an appropriate orthogonal array table is selected. Furthermore, it is preferable to select such that the statistical reliability is 95% and the detection power is 80% or more.

In other words, the combination determining unit 332 performs at least one of the following (1) to (3).
(1) Select at least two modules M that have been determined to have a high effect measure value in the past (2) Select element data E that has been determined to have a high effect measure value in the past (3) Present Non-all combinations and their numbers based on the advertising traffic of

3. Advertising Effect Verification Method In section 3, an example of an advertisement effect verification method using the above-described system 1 will be described. FIG. 7 is a flowchart showing a method of verifying the effect of an advertisement, and will be described in detail below along each step in FIG.

[start]
(Step S1)
The user U uses the input unit 25 in the user terminal 2 to input the original advertisement data OA. It should be noted that the user U may create the original advertisement data OA at the user terminal 2, or the user U may download the advertisement data OA from an external advertisement database to the user terminal 2 via a network. The original advertisement data OA stored in the storage unit 22 of the user terminal 2 in this way is transmitted to the information processing device 3 via the communication unit 21 and the network (following step S2).

(Step S2)
The communication unit 31 in the information processing device 3 receives the original advertisement data OA transmitted from the user terminal 2 in step S1, and this is stored in the storage unit 32 in the information processing device 3. Subsequently, the advertisement analysis unit 330 in the control unit 33 analyzes the original advertisement data OA stored in the storage unit 32. Specifically, the original advertisement text data OAT as text data and the original advertisement image data OAI as image data are decomposed, and element data E is extracted from each. Here, among text data, image data, vector data, effect information, or audio data, element data E composed of text data and image data is described. Then, for each element data E, new element data Ea is proposed based on a rule defined by at least one of the appearance, name, and idea of the element data E, and the element data E, Ea based on the rule are proposed. Are defined as a module M. (Subsequent to step S3).

(Step S3)
Subsequent to step S2, the combination determining unit 332 of the control unit 33 determines, among the combinations of the module M and the element data E that are candidates, the combinations to be actually verified based on the orthogonal array table in statistics. To determine. Further, the combination determination unit 332 ranks the non-all combinations in accordance with a predetermined rule (following step S4).

(Step S4)
Subsequent to step S3, the advertisement generation unit 333 of the control unit 33 generates a plurality of pieces of advertisement data corresponding to the combination determined in step S3. Note that one of the plurality of generated advertisement data may be the same as the original advertisement data OA. Further, the original advertisement data OA received in step S2 may be used in the first place. In particular, it is preferable that at least two pieces of advertisement data to be verified at the same time are generated under the ranking determined in step S3 (following step S5).

(Step S5)
Following step S4, the effect verification unit 334 of the control unit 33 verifies the effect of the advertisement for at least two pieces of advertisement data to be verified generated in step S4 in a predetermined period. At the time of verification, if there is an advertisement whose effect measure does not satisfy the threshold value even within the period, the advertisement is stopped, or at least one advertisement whose effect measure does not satisfy the threshold value. Note that if any, all ads being verified at the same time will be stopped.

(Step S6)
By repeating the processes in steps S4 and S5 as necessary until the non-all combinations are completed, the effect of the advertisement is finally verified. Then, the verification result R is transmitted to the user terminal 2 via the communication unit 31 in the information processing device 3 and the network. The user U can browse the result on the display unit 24 of the user terminal 2. Further, it is preferable that the verification result R includes not only information on the verification results regarding at least two advertisements described above but also information regarding an optimal advertisement based on the results. The information on the optimal advertisement may be one of the advertisements related to the non-all combinations, or may be another advertisement.

FIG. 10 shows an example of the verification result R. An analysis report obtained by performing multivariate analysis for each element data E for a certain module M can be obtained as the verification result R. Thereby, the user U can grasp the effective element data E.
[End]

4. Effects According to the system 1 according to the present embodiment, the following various effects can be obtained.

  According to the system 1 according to the present embodiment, the guide information G allows the user U to intuitively select an efficient orthogonal array table without having knowledge of statistics. That is, an optimal test plan can be proposed.

  In addition, by explicitly indicating the reduction efficiency, the efficiency of the verification is specified, and a more preferable combination can be selected.

  The proposal by the system 1 can be reflected with one button, and the work load and the psychological load for the change can be reduced.

  Since the proposal is changed to the optimal proposal in real time according to the current selection, the optimal experiment plan (orthogonal array table) can be adopted according to the situation.

  Since the error information is displayed, it is possible to judge instantly whether or not there is a problem. In addition, since the error is specified at the place where the error is made, it is easy to grasp where to correct.

  Since an optimal orthogonal array table is proposed in consideration of errors, it is possible to respect changes in the thinking of the user U and always present preferable options.

  Since the same preferred experimental design can be assigned without repeating the work, it is expected that the enormous work time of the creative creative creator will be reduced.

  It is possible to obtain a multivariate analyzed analysis report for each element data E with minimum information. That is, it is possible to reduce the label assignment work for the analysis and to generate an advanced analysis report.

  When conducting comparative tests (A / B test, split run test) on ads that are placed in online advertising or text, images, or a combination of them, generate a large amount of ads with minimum effort And easily perform the analysis. In addition, it is possible to analyze both a text advertisement using a combination of characters and an image image advertisement using a combination of images on a disassembled unit basis.

  Furthermore, by supporting the test of the text advertisement in units of words and phrases, it is possible to generate dozens of patterns of test target advertisements by, for example, performing a few clicks on the display screen. It is possible to provide a situation from test execution to tallying, regardless of literacy.

  In other words, in order to perform word analysis and recommendation, when a word is selected, an ad text corresponding to the selected word is automatically generated, and when a campaign or ad group to be registered is specified, the system automatically submits an advertisement via an API. The results are totaled every day, and when the scheduled completion time is reached, the end processing is performed and the report is output, so that the user of the advertisement proposal system hardly takes time.

  Also, at the time of generating a text advertisement, a test pattern is coded (for example, in the case of a test of 0-1-0-3, 4 phrases), recorded, and recorded as test design data, so that all tests can be performed. It can be easily analyzed. In addition, in order to generate a list (test design list) of which word (element data E) corresponds to which digit and in what number, what kind of test was performed based on this design document Tracing becomes possible.

  Furthermore, with the function of generating an image advertisement by combining image parts (element data E), various combinations can be generated instantaneously, and at the same time, text input by the user U can be generated as an image. Thus, the user U can execute the test of the image advertisement in the same manner as the text advertisement without relying on the creative staff.

  Further, a pattern indicating which element data E is combined is coded (for example, in the case of a 4-part test of 0-1-0-3) and incorporated into the image name, whereby the test design data and the image name are combined. If it is prepared, the results for each element data E can be totaled at any time.

  Furthermore, since the test pattern can be re-extracted at any time by the coding described above, the past data can be instantaneously subjected to the multiple regression analysis at any time based on the test pattern. If this process does not exist, a work of disassembling words and parts is required before the totaling, and the load until analysis is high. However, by this process, the totaling can be freely performed at any time.

  Furthermore, by totalizing the above-mentioned tests for each user to be advertised, it becomes possible to grasp the test execution frequency, winning rate, and improvement rate for each user to be advertised in real time.

  Further, by setting sharing within an organization such as a company, it is possible to browse past execution results on a part-by-part basis and share the results as knowledge. That is, in the past, the test was often performed as a completed advertising text, so that the test result could not be reused, and it was more difficult to utilize the knowledge for other projects and other users. With this system, knowledge can be created because it is decomposed into words and image parts.

  Further, machine learning is performed based on the action of the advertising target user in the organization, and the recommendation can be performed based on the behavior selected by the advertising target user in the past. In other words, in order to accumulate all the performance data and test design work divided into knowledgeable units in one system, each action of the advertising target user and each test result are taken in as learning data. It can be used for general purposes.

  In addition, since the creation of the test, publication, monitoring, and analysis are automated, the PDCA cycle of the test can be performed without the manual work of the user of the advertisement proposal system.

  In particular, as described in Section 3, the processes related to steps S4 and S5 are repeated as necessary until the non-all combinations are completed, and in this process, the combination ( Since the winning advertisement is always kept, it is possible to trace the continuous improvement of the campaign effect in a plurality of test executions, instead of repeating a single improvement.

  Further, the effect verification unit 334 sets a threshold value (number of days, number of display times, generated volume, and the like) necessary for evaluation, thereby making it possible to avoid erroneous determination based on evaluation with low reliability.

  Further, in the process of repeated execution by the effect verification unit 334, a fixed forced stop rule can be executed in parallel, so that excessive performance deterioration due to the new element data E can be prevented.

  In addition, a mechanism is provided for preventing erroneous determination based on data of different periods between advertisements in the process of repeated execution by the effect verification unit 334. For example, when advertisements A, B, and C are placed, and A is an existing advertisement, and B and C are additional advertisements, the start date of B and C is determined by the timing of examination on the medium side in the advertisement operation. Deviation can occur.

  For example, if A has data for 7 days, B has data for 5 days, and C has data for 3 days, if all are compared with data for 7 days, the concept of the day of the week may be ignored and evaluation between advertisements may be performed. Since this is apt to cause an evaluation error, the advertisement is evaluated only in the minimum period. That is, A vs B compares data for 5 days, and A vs C compares data for 3 days. Because of this mechanism, it is possible to release the inefficient method of starting the test after confirming the start of the advertisement every time, and to appropriately detect the difference between the advertisements.

5. Conclusion As described above, according to the present embodiment, it is possible to implement the information processing device 3 that can efficiently verify the effect of an advertisement without being conscious of statistics.

  That is, an information processing apparatus for generating a plurality of advertisement candidates that are candidates for a desired advertisement, wherein data constituting the advertisement is defined as element data, and the element data is defined as an appearance, a name, and an idea. When a module classified and summarized based on a rule defined by at least one of the following is defined as a module, the information processing apparatus includes a storage unit, a guide information generation unit, a communication unit, and an advertisement generation unit, The storage unit stores the element data in association with the module, and the guide information generation unit is configured to be able to generate at least one guide information, wherein the guide information is the element for each of the modules. The total number of combinations of data, the number of verifications verified by the orthogonal array table, and the plurality of selectable element data associated with the module are listed one by one. The communication unit is configured to be able to transmit the guide information to the user terminal of the user, configured to be able to receive the orthogonal array table selected by the user from the user terminal, the advertisement generation unit The present invention provides an information processing apparatus configured to generate an advertisement candidate having a verification number specified in the orthogonal array table.

  In addition, software for implementing the information processing device 3 capable of efficiently verifying the effect of an advertisement without being conscious of statistics may be implemented as a program. Then, such a program may be provided as a non-transitory recording medium readable by a computer, may be provided so as to be downloadable from an external server, or may be started by an external computer. In addition, so-called cloud computing that can execute each function on the client terminal may be performed.

  That is, the program is a program that causes a computer to function as an information processing apparatus for generating a plurality of advertisement candidates that are candidates for a desired advertisement, defines data constituting the advertisement as element data, and displays the element data in appearance. When a module classified and summarized based on a rule defined by at least one of a name and an idea is defined as a module, the information processing apparatus includes a storage unit, a guide information generation unit, a communication unit, and an advertisement generation unit. The storage unit stores the element data in association with the module, and the guide information generation unit is configured to be able to generate at least one guide information, wherein the guide information is the module The total number of combinations of the element data for each, the number of verifications verified by the orthogonal array table, and the selectable number associated with the module. A table in which the element data are listed one by one, the communication unit is configured to be able to transmit the guide information to a user terminal of the user, and the orthogonal array table selected by the user from the user terminal. Is provided, and the advertisement generation unit generates an advertisement candidate having a verification number specified in the orthogonal array table.

  Further, it is possible to implement the system 1 capable of efficiently verifying the effect of the advertisement without paying attention to statistics.

  That is, the system is for generating a plurality of advertisement candidates that are candidates for a desired advertisement, includes an information processing device and a user terminal, and defines data constituting the advertisement as element data, When the element data is classified and grouped based on a rule defined by at least one of appearance, name, and idea, and defined as a module, the information processing apparatus includes a storage unit, a guide information generation unit, A communication unit and an advertisement generation unit, wherein the storage unit stores the element data in association with the module, and the guide information generation unit is configured to generate at least one guide information. In the guide information, the total number of combinations of the element data for each module, the number of verifications verified by an orthogonal array table, and the selection linked to the module And a table in which the plurality of element data items are listed one by one, wherein the first communication unit is configured to be able to transmit the guide information to a user terminal of the user, and the user terminal The selected orthogonal arrangement table is configured to be receivable, the advertisement generation unit is configured to be able to generate an advertisement candidate of a verification number defined in the orthogonal arrangement table, the user terminal includes a display unit, an input unit, , A second communication unit, the display unit displays guide information received from the information processing device, the input unit is configured to be able to accept an orthogonal array selected by a user, the A communication system is provided, wherein the second communication unit is configured to be able to transmit the selected orthogonal array table to the information processing apparatus.

  Finally, various embodiments according to the invention have been described, but these have been presented by way of example and are not intended to limit the scope of the invention. The new embodiment can be implemented in other various forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. The embodiment and its modifications are included in the scope and gist of the invention, and are also included in the invention described in the claims and equivalents thereof.

1: system 2: user terminal 20: communication bus 21: communication unit 22: storage unit 23: control unit 24: display unit 25: input unit 3: information processing device 30: communication bus 31: communication unit 32: storage unit 33: Control unit 330: advertisement analysis unit 331: guide information generation unit 332: combination determination unit 333: advertisement generation unit 334: effect verification unit BI: background image E: element data E1: first element data E1a: first element data E2: second element data E2a: second element data E3: third element data E3a: third element data E4: fourth element data E4a: fourth element data EP: description image Ea: Element data G: guide information LG: logo mark M: module M1: first module M2: second module M3: third module M4: fourth module MI: Main image OA: Original advertisement data OAI: Original advertisement image data OAT: Original advertisement text data R: Verification result SI: Sub-image T: Table TL: Title image U: User

Claims (11)

An information processing apparatus for generating a plurality of advertisement candidates that are candidates for a desired advertisement,
Defining the data that constitutes the advertisement as element data, and classifying and summarizing the element data based on rules defined by at least one of appearance, name, and idea as a module,
The information processing device includes a storage unit, a guide information generation unit, a communication unit, and an advertisement generation unit,
The storage unit stores the element data in association with the module,
The guide information generation unit is configured to generate at least one guide information, wherein the guide information includes a total number of combinations of the element data for each module and a verification number verified by an orthogonal array table. And a table in which a plurality of selectable element data associated with the module are listed one by one, including information that can be confirmed with a list ,
The communication unit,
The guide information is configured to be transmitted to a user terminal of a user,
From the user terminal, configured to be able to receive the orthogonal arrangement table selected by the user,
The advertisement generation unit is configured to be able to generate an advertisement candidate of a verification number defined in the orthogonal array table,
Information processing device. The information processing apparatus according to claim 1,
When a measure using at least one of the number of inflows by clicks, the number of queries, the unit price of induction, and the unit price of acquisition for the advertisement as variables is defined as an effect measure,
The information processing apparatus further includes a combination determination unit,
The storage unit stores a past effect measure related to the advertisement candidate,
The combination determining unit selects at least two of the modules in which the value of the effect measure has been determined to be high in the past and the element data included in the modules and in which the value of the effect measure has been determined to be high in the past. In this way, non-all combinations and the number of the element data are determined based on the orthogonal array table,
Information processing device. The information processing apparatus according to claim 2,
Information processing device.
It further includes an effect verification unit,
The effect verification unit, among the advertisements having the verification number,
If there is an advertisement candidate whose effectiveness measure does not satisfy the threshold, the verification of the advertisement candidate is stopped, or if there is even one advertisement candidate whose effectiveness measure does not satisfy the threshold, all the advertisement candidates are checked. Stop validation,
By repeating the generation of the advertisement candidate by the advertisement generation unit and the verification of the advertisement candidate by the effect verification unit, it is configured to be able to verify the effect of the desired advertisement,
Information processing device In the information processing device according to claim 2 or 3,
The effect verification unit is configured to perform a multivariate analysis on the advertisement candidates of the verification number, and to generate effect data indicating an effect for each of the element data,
The communication unit is configured to be able to transmit the effect data to a user terminal of the user,
Information processing device. The information processing device according to claim 3,
There are a plurality of each of the advertisement generation unit and the effect verification unit,
Configured to verify the effectiveness of the advertisement by parallel operation;
Information processing device. In the information processing apparatus according to any one of claims 2 to 5,
The combination determination unit is configured to determine the non-all combinations and to be able to perform ranking of these combinations.
Information processing device. The information processing apparatus according to claim 6,
The combination determination unit is configured to be able to perform the ranking in the order that the effect of the advertisement is predicted to be high from the verification result already performed in the past, In the information processing apparatus according to any one of claims 1 to 7,
It further has a warning section,
The warning unit is configured to be able to generate warning information according to at least one of the following (1) to (4) when the orthogonal array table selected by the user is not suitable,
(1) The number of the modules or the element data is excessive (2) The number of the modules or the element data is insufficient (3) The module or the element data candidates to be deleted (4) The module or the candidate of the element data to be added The communication unit is configured to be able to transmit the warning information to the user terminal,
Information processing device. In the information processing apparatus according to any one of claims 1 to 8,
The element data includes text data, image data, vector data, effect information, or audio data,
Information processing device. A program that causes a computer to function as an information processing device for generating a plurality of advertisement candidates that are candidates for a desired advertisement,
Defining the data that constitutes the advertisement as element data, and defining a module that classifies and summarizes the element data based on rules defined by at least one of appearance, name, and idea, as a module,
The information processing device includes a storage unit, a guide information generation unit, a communication unit, and an advertisement generation unit,
The storage unit stores the element data in association with the module,
The guide information generation unit is configured to generate at least one guide information, wherein the guide information includes a total number of combinations of the element data for each module and a verification number verified by an orthogonal array table. And a table in which a plurality of selectable element data associated with the module are listed one by one, including information that can be confirmed with a list ,
The communication unit,
The guide information is configured to be transmitted to a user terminal of the user,
From the user terminal, configured to be able to receive the orthogonal arrangement table selected by the user,
The advertisement generation unit generates an advertisement candidate having a verification number defined in the orthogonal array table,
program. A system for generating a plurality of advertisement candidates that are desired advertisement candidates,
Comprising an information processing device and a user terminal,
Defining the data that constitutes the advertisement as element data, and defining a module that classifies and summarizes the element data based on rules defined by at least one of appearance, name, and idea, as a module,
The information processing device includes a storage unit, a guide information generation unit, a first communication unit, and an advertisement generation unit,
The storage unit stores the element data in association with the module,
The guide information generation unit is configured to generate at least one guide information, wherein the guide information includes a total number of combinations of the element data for each module and a verification number verified by an orthogonal array table. And a table in which a plurality of selectable element data associated with the module are listed one by one, including information that can be confirmed with a list ,
The first communication unit includes:
The guide information is configured to be transmitted to a user terminal of the user,
From the user terminal, configured to be able to receive the orthogonal arrangement table selected by the user,
The advertisement generation unit is configured to be able to generate an advertisement candidate of a verification number defined in the orthogonal array table,
The user terminal includes a display unit, an input unit, and a second communication unit,
The display unit displays guide information received from the information processing device,
The input unit is configured to be able to receive an orthogonal array table selected by a user,
The second communication unit is configured to be able to transmit the selected orthogonal array table to the information processing device.
system.