Evaluation device, evaluation method, and computer program

Abstract

Improve the accuracy of video ad ratings. [Solution] An evaluation device for evaluating video advertisements includes an acquisition unit and an evaluation unit. The acquisition unit is configured to acquire features related to one or more expression elements included in the video advertisement. The evaluation unit is configured to input the features into a trained machine learning model, and acquire from the machine learning model an evaluation value for the video advertisement that the machine learning model outputs based on the features using one or more evaluation indexes. The one or more expression elements include at least one of visual information and auditory information. The one or more evaluation indexes include one or more indexes related to advertising effectiveness.

Description

[Technical Field]

[0001] The present disclosure relates to an evaluation device. [Background technology]

[0002] Techniques for evaluating advertisements are known. For example, Patent Document 1 discloses an information processing program for evaluating documents used in advertisements and improving the components of the advertisements. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Application Laid-Open No. 2011-128761 Summary of the Invention [Problem to be solved by the invention]

[0004] Video advertisements such as television commercials and web advertisements contain a wide variety of expression elements. For this reason, it is difficult to evaluate video advertisements with high accuracy by simply considering the documents used in the advertisements, as in the information processing program disclosed in Patent Document 1.

[0005] One aspect of the present disclosure is to improve the accuracy of video advertisement evaluation. [Means for solving the problem]

[0006] One aspect of the present disclosure is an evaluation device for evaluating a video advertisement, comprising an acquisition unit and an evaluation unit. The acquisition unit is configured to acquire feature quantities related to one or more expression elements included in the video advertisement. The evaluation unit is configured to input the feature quantities into a trained machine learning model, and acquire from the machine learning model an evaluation value for the video advertisement output by the machine learning model using one or more evaluation indexes based on the feature quantities. The one or more expression elements include at least one of visual information and auditory information. The one or more evaluation indexes include one or more indexes related to advertising effectiveness.

[0007] According to this configuration, the video advertisement can be evaluated using an evaluation value based on at least one of visual information and audio information included in the video advertisement, thereby improving the accuracy of the evaluation of the video advertisement.

[0008] In one aspect of the present disclosure, one or more expressions may include auditory information. In one aspect of the present disclosure, one or more expressions may include both visual and auditory information.

[0009] According to this configuration, the video advertisement can be evaluated using the evaluation index based on the auditory information included in the video advertisement, thereby improving the accuracy of the evaluation of the video advertisement.

[0010] In one aspect of the present disclosure, the acquisition unit may be configured to acquire, as the feature, appearance times of each of the one or more expression elements in the video advertisement, and the evaluation unit may be configured to input, as the feature, the appearance times corresponding to each of the one or more expression elements into the machine learning model.

[0011] In one aspect of the present disclosure, the appearance time may be, for each of one or more expression elements, the total amount of time that the corresponding expression element appears in the video advertisement. According to this configuration, it is possible to evaluate a video advertisement not only based on the expression elements but also based on the appearance time of the expression elements, thereby improving the accuracy of the evaluation of the video advertisement.

[0012] According to an aspect of the present disclosure, the system may further include a calculation unit. The one or more expression elements may include a plurality of elements. The calculation unit may be configured to calculate a contribution level, which is a degree to which each of the plurality of elements contributes to each of the one or more evaluation indexes.

[0013] According to this configuration, the degree of contribution of each expression element included in the video advertisement can be used for evaluation. In one aspect of the present disclosure, the calculation unit may be configured to identify at least one element having a relatively high degree of contribution from among the plurality of elements.

[0014] With this configuration, it is possible to identify which expression elements in a video advertisement contribute to the advertisement's effectiveness, thereby improving the video advertisement. One aspect of the present disclosure may be a computer-implemented evaluation method for evaluating a video advertisement. The evaluation method includes acquiring features related to one or more expression elements included in the video advertisement, inputting the features into a trained machine learning model, and acquiring, from the machine learning model, an evaluation value for the video advertisement output by the machine learning model using one or more evaluation indicators based on the features. The one or more expression elements include at least one of visual information and auditory information. The one or more evaluation indicators include one or more indicators related to advertising effectiveness.

[0015] This evaluation method provides the same effects as the above-mentioned evaluation device. In one aspect of the present disclosure, a computer program may be provided for causing a computer to execute at least part of the above-described evaluation method. [Brief explanation of the drawings]

[0016] [Figure 1] FIG. 1 is a block diagram showing a configuration of an evaluation system. [Figure 2] FIG. 1 is a diagram illustrating the input and output of a machine learning model. [Figure 3] 10 is a diagram illustrating the relationship between a group of past advertisements stored in a database, a list of expression elements, and evaluation data. FIG. [Figure 4] FIG. 10 is a diagram illustrating expression elements and appearance times in the expression element list. [Figure 5] FIG. 10 is a diagram illustrating the total values ​​of expression elements and appearance times in the expression element list. [Figure 6] 10 is a flowchart illustrating a learning process. [Figure 7]10 is a flowchart showing an evaluation process. [Figure 8] 10 is an example of a screen showing a result of the evaluation process. [Figure 9] FIG. 10 is a diagram illustrating an output of a machine learning model according to another embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0017] Hereinafter, exemplary embodiments of the present disclosure will be described with reference to the drawings. [1. First embodiment] [1-1.Configuration] [1-1-1. Overall structure] The evaluation system 100 shown in FIG. 1 is a system for evaluating video advertisements broadcast on media such as television and the Internet.

[0018] The evaluation system 100 includes an evaluation device 1, a machine learning model 2, and a database 3. The evaluation device 1 includes a processor 11, a memory 12, a storage 13, a user interface 14, and a communication interface 15. The evaluation device 1 is installed in an information terminal such as a personal computer.

[0019] The processor 11 is configured to execute processing in accordance with a computer program recorded in the storage 13 . The memory 12 is used as a work area when the processor 11 executes processing. Examples of the memory 12 include a random access memory (RAM), a read only memory (ROM), and a flash memory.

[0020] The storage 13 holds computer programs and data used when executing processes according to the computer programs. Examples of the storage 13 include a hard disk drive (HDD) and a solid state drive (SSD).

[0021] The user interface 14 is a general term for an interface for receiving various input operations from a user and an interface for outputting various information to the user. Examples of the user interface 14 include a keyboard, a mouse, a touch panel, a display, etc.

[0022] The communication interface 15 is an interface capable of communicating various types of data in accordance with a predetermined standard. The evaluation device 1 is configured to be able to communicate with the machine learning model 2 and the database 3 via the communication interface 15.

[0023] The machine learning model 2 is a model that has been trained using machine learning. The machine learning model 2 may be a model that has been trained using deep learning. As shown in Figure 2, when feature quantities 41 related to one or more expression elements 40 included in a video advertisement are input, the machine learning model 2 is configured to output an evaluation value 51 calculated using one or more evaluation indicators 5 for the video advertisement based on the feature quantities 41, and a contribution degree 52 corresponding to each of the one or more expression elements 40.

[0024] The one or more expression elements 40 include at least one of visual information and audio information contained in the video advertisement. Visual information is information that appeals to viewers of the video advertisement through their sense of sight. Examples of visual information include people such as celebrities and characters, and text such as catchphrases and subtitles. Audio information is information that appeals to viewers of the video advertisement through their sense of hearing. Examples of audio information include sounds such as dialogue and narration, music such as songs and background music, and sound effects.

[0025] The expression element 40 may be a performance or expression. For example, the expression element 40 may be a sizzle that appeals to the sensory appeal of a product through sight and sound. Examples of sizzle include an ingredient sizzle in an advertisement for food ingredients or cooking, which displays the ingredients themselves along with sound effects, and a cooking sizzle that includes the appearance and sounds of cooking, such as cutting or frying ingredients.

[0026] The one or more evaluation indexes 5 are indexes relating to the advertising effectiveness of the video advertisement. An example of the evaluation index 5 is an index relating to the advertising expression effectiveness. Examples of the index relating to the advertising expression effectiveness include an attractiveness presentation level, which indicates the degree to which the appeal of a brand, product, etc. is presented (i.e., how attractively the brand, product, etc. was presented), and an attention arousal level, which indicates the degree to which attention is drawn to the video advertisement (i.e., how conspicuous the advertising expression was).

[0027] The higher the attractiveness level, the more likely the video advertisement will make the product seem attractive to viewers. The higher the level of attention-grabbing, the more prominent the advertisement in the video advertisement will be, and the more likely it will attract the viewer's attention.

[0028] The evaluation value 51 is a value obtained by evaluating a video advertisement using the evaluation index 5. The evaluation value 51 is calculated for each of one or more evaluation indexes 5. The evaluation value 51 may be a numerical value obtained by evaluating the video advertisement, or may be a rank obtained by evaluating the advertisement in stages based on the numerical value. As an example, FIG. 2 shows the evaluation values ​​51 for a certain video advertisement, which correspond to the attractiveness level and attention-grabbing level as evaluation indexes 5, as numerical values ​​of "94.1" and "86.5", respectively.

[0029] As shown in FIG. 3, the database 3 includes a group of past advertisements ADG, a list of expression elements IL, and evaluation data EV. The past advertisement group ADG includes multiple video advertisements that have been broadcast in the past.

[0030] The expression element list IL has a list including feature amounts 41 relating to one or more expression elements 40 corresponding to each of the video advertisements in the past advertisement group ADG. The evaluation data EV includes one or more evaluation values ​​51 corresponding to each of the video advertisements in the past advertisement group ADG.

[0031] An example of evaluation data EV is Best HIT (registered trademark). Best HIT is a fixed-point observation survey that measures the advertising effectiveness of television commercials broadcast weekly on Tokyo's key broadcasting stations. Best HIT contains data evaluating each television commercial based on indicators related to the effectiveness of advertising presentation, including the degree of attractiveness and attention-grabbing ability.

[0032] The contribution degree 52 is the degree to which each of one or more expression elements 40 contributes to an evaluation index 5. In a certain video advertisement, an expression element 40 having a high contribution degree 52 to a certain evaluation index 5 has a high degree of contribution to the evaluation index 5 of the video advertisement.

[0033] The contribution degree 52 is calculated for each evaluation index 5. For example, when a first evaluation index 5a and a second evaluation index 5b are used as one or more evaluation indexes 5, a first contribution degree 52a as the contribution degree 52 to the first evaluation index 5a and a second contribution degree 52b as the contribution degree 52 to the second evaluation index 5b are calculated.

[0034] For example, if the first evaluation index 5a is the attractiveness level, the first contribution 52a represents the degree to which the corresponding expression element 40 contributes to the attractiveness level. If the second evaluation index 5b is the attention arousal level, the second contribution 52b represents the degree to which the corresponding expression element 40 contributes to the attention arousal level.

[0035] In the example of FIG. 2, for "expression element B," the "contribution to attractiveness presentation" as the first contribution 52a is "0.556." The "contribution to attention arousal" as the second contribution 52b is "-0.119." In other words, even if the contribution 52 to one evaluation index 5 is high for the same expression element 40, the contribution 52 to another evaluation index 5 may be low.

[0036] [1-1-2. List of expression elements] As shown in FIG. 4, the expression element list IL includes, as feature quantities 41 related to one or more expression elements 40 in one video advertisement, appearance times 41a of each of one or more expression elements 40 in the video advertisement.

[0037] The appearance time 41a indicates the time period during which the expression element 40 appears in the video advertisement. Specifically, the appearance time 41a indicates the start time and end time when the expression element 40 appears in the video advertisement. The start time and end time are times based on the beginning of the video advertisement. The appearance time 41a may be expressed by the start time and elapsed time instead of the start time and end time. The appearance time 41a may be continuous or may not be continuous.

[0038] In this embodiment, as an example, the appearance time 41a indicates the start and end times at which the representation element 40 appears in a 15-second video advertisement, expressed in units of 0.1 seconds. For example, the appearance time 41a of the representation element 40 "talent" is the time period between 0.5 seconds and 5.2 seconds and between 8.4 seconds and 15.0 seconds after the start of the video advertisement.

[0039] 4, the shaded area for each expression element 40 visually represents the appearance time 41a. The numbers 1 to 15 above the shaded area represent the number of seconds (rounded up to the nearest whole number) that have elapsed since the start of the video advertisement.

[0040] 5, the appearance time 41a may be the total time that each of one or more expression elements 40 appears in the video advertisement. In the above example, the appearance time 41a of the expression element 40 "talent" is 11.3 seconds.

[0041] The length of the video advertisement is not limited to 15 seconds, but may be, for example, 30 seconds, 60 seconds, etc. The unit of appearance time 41a is not limited to 0.1 seconds, but may be, for example, 0.5 seconds, 1 second, etc.

[0042] The expression elements 40 and appearance times 41a may be associated with each video advertisement manually by a user, or may be associated automatically by a computer capable of analyzing the content of the video.

[0043] [1-2. Processing] [1-2-1. Learning process] In the evaluation system 100, the evaluation device 1 is configured to execute the learning process shown in FIG.

[0044] The learning process is a process of generating a machine learning model 2 through machine learning. The learning process executed by the processor 11 of the evaluation device 1 will be described below with reference to the flowchart of FIG.

[0045] When an instruction to execute the learning process is input from the user via the user interface 14, the processor 11 starts the process shown in FIG. First, in S100, the processor 11 acquires the past advertisement group ADG, the expression element list IL, and the evaluation data EV from the database 3 via the communication interface 15.

[0046] Next, in S110, the processor 11 generates a machine learning model 2 by performing supervised learning. The machine learning model 2 has, as explanatory variables, feature quantities 41 related to one or more expression elements 40. In this embodiment, the machine learning model 2 has, as feature quantities 41 corresponding to each of the one or more expression elements 40, explanatory variables that indicate appearance times 41a of the corresponding expression elements 40.

[0047] The machine learning model 2 has one or more evaluation indicators 5 as objective variables. In supervised learning, the explanatory variables are assigned features 41 included in the expression element list IL corresponding to each video advertisement included in the past advertisement group ADG. The objective variable is assigned evaluation data EV of the corresponding video advertisement. The machine learning model 2 is generated based on combinations of multiple expression elements 40 included in the expression element list IL, combinations of expression elements 40 and appearance times 41a, etc.

[0048] Thereafter, the processor 11 ends the processing shown in FIG. In this way, the processor 11 performs machine learning using various data acquired from the database 3 and generates the machine learning model 2.

[0049] [1-2-2. Evaluation process] In the evaluation system 100, the evaluation device 1 is configured to execute the evaluation process shown in FIG.

[0050] The evaluation process is a process of evaluating a video advertisement using the machine learning model 2. In other words, the evaluation process is a process of predicting an evaluation value 51 of a video advertisement using the machine learning model 2. The evaluation process executed by the processor 11 of the evaluation device 1 will be described below with reference to the flowchart of FIG.

[0051] When an instruction to execute the evaluation process is input from the user via the user interface 14, the processor 11 starts the process shown in FIG. First, in S200, the processor 11 acquires feature quantities 41 related to one or more expression elements 40 in the video advertisement to be evaluated. The processor 11 may acquire the feature quantities 41 from the storage 13, or may acquire the feature quantities 41 from outside the evaluation device 1 (for example, from the database 3) via the communication interface 15. In this embodiment, the feature quantities 41 include appearance times 41a corresponding to each of the one or more expression elements 40.

[0052] The feature quantity 41 in the video advertisement to be evaluated may be extracted manually in advance by a user, or may be extracted automatically in advance by a computer capable of analyzing the content of the video.

[0053] Next, in S210, the processor 11 inputs the feature quantities 41 related to the one or more expression elements 40 acquired in S200 into the machine learning model 2, and obtains an evaluation value 51 as an output of the machine learning model 2 based on the feature quantities 41, thereby predicting the evaluation value 51 of the video advertisement to be evaluated. The processor 11 displays the acquired evaluation value 51 to the user via the user interface 14.

[0054] Subsequently, in S220, the processor 11 uses the output of the machine learning model 2 in S210 to calculate a contribution 52, which is the degree of contribution of the expression element 40 to the evaluation index 5. When the processor 11 inputs the feature quantities 41 related to the plurality of expression elements 40 to the machine learning model 2 in S210, it calculates the contribution 52 for each expression element 40. The contribution 52 is calculated using, for example, SHAP (SHapley Additive exPlanations).

[0055] Next, in S230, the processor 11 identifies at least one expression element 40, of the expression elements 40 relating to the feature quantity 41 input to the machine learning model 2, that has a relatively high contribution degree 52 calculated in S220.

[0056] For example, the processor 11 identifies one expression element 40 with the highest contribution degree 52 among the expression elements 40 related to the feature quantity 41 input to the machine learning model 2. Alternatively, the processor 11 identifies a predetermined number of expression elements 40 in descending order of contribution degree 52.

[0057] According to another example, the processor 11 identifies at least one expression element 40 having a contribution degree 52 exceeding a predetermined threshold value from among the expression elements 40 related to the feature quantity 41 input to the machine learning model 2. For example, the processor 11 identifies, from among the expression elements 40 related to the feature quantity 41 input to the machine learning model 2, an expression element 40 whose contribution degree 52 is a positive value.

[0058] The processor 11 displays the identified expression element 40 and the contribution degree 52 corresponding to the expression element 40 to the user via the user interface 14. 8 is an example of a screen displaying, as a result of the above process, the evaluation value 51 of the attractiveness level as the evaluation index 5 and the top three expression elements 40 with the highest contribution rate 52 for the video advertisement to be evaluated on the display of the user interface 14. FIG. 8 also shows the bottom two expression elements 40 with the lowest contribution rate 52.

[0059] Thereafter, the processor 11 ends the processing shown in FIG. In this way, the processor 11 is configured to input the feature quantities 41 related to the expression elements 40 of the video advertisement to be evaluated to the machine learning model 2, obtain an evaluation value 51 as an output from the machine learning model 2, and evaluate the video advertisement to be evaluated. Furthermore, the processor 11 is configured to calculate a contribution degree 52 to the evaluation index 5 corresponding to the obtained evaluation value 51 for each expression element 40, and identify expression elements 40 with a relatively high contribution degree 52.

[0060] [1-3.Effects] According to the embodiment described above, the following actions and effects can be obtained. (1a) In the learning process, the processor 11 generates a machine learning model 2 by performing supervised learning. The machine learning model 2 has feature quantities 41 related to one or more expression elements 40 as explanatory variables. The machine learning model 2 has one or more evaluation indicators 5 as objective variables. In supervised learning, the explanatory variables are assigned a list of expression elements IL corresponding to each video advertisement included in the past advertisement group ADG. The objective variable is assigned evaluation data EV of the corresponding video advertisement.

[0061] In the evaluation process, the processor 11 inputs features 41 related to one or more expression elements 40 in the video advertisement to be evaluated into the machine learning model 2, and obtains an evaluation value 51 as an output of the machine learning model 2 based on the features 41, thereby predicting the evaluation value 51 of the video advertisement to be evaluated.

[0062] The expression element 40 includes at least one of visual information and audio information included in the video advertisement. According to this process, the video advertisement can be evaluated based on at least one of the visual information and the audio information included in the video advertisement, using the evaluation value 51. This makes it possible to improve the accuracy of the evaluation of the video advertisement.

[0063] (1b) The machine learning model 2 can have, as the feature 41 corresponding to each of one or more expression elements 40, an explanatory variable that indicates the appearance time 41a of the corresponding expression element 40. According to this process, it is possible to evaluate the video advertisement in consideration of the appearance time 41a of the expression element 40. This makes it possible to improve the accuracy of the evaluation of the video advertisement.

[0064] (1c) In the evaluation process, the processor 11 uses the output of the machine learning model 2 to calculate the contribution 52, which is the degree of contribution of the expression element 40 to the evaluation index 5. According to this process, the contribution degree 52 of each of the expression elements 40 included in the video advertisement can be used for evaluation.

[0065] (1d) The processor 11 identifies, from among the expression elements 40 relating to the feature quantity 41 input to the machine learning model 2, the expression elements 40 whose calculated contribution degree 52 is relatively high. This process makes it possible to identify which expression elements 40 contribute to the effectiveness of a video advertisement. Therefore, for example, a user can improve a video advertisement that is currently being produced by selecting and discarding expression elements 40 or adjusting the appearance time 41a using the contribution degree 52. Specifically, the user can improve the video advertisement by lengthening the appearance time 41a of an expression element 40 with a high contribution degree 52 or by replacing an expression element 40 with a low contribution degree 52 with another expression element 40.

[0066] [1-4. Correspondence between terms] In the above embodiment, the processing of S200 corresponds to an example of processing executed as an acquisition unit, the processing of S210 corresponds to an example of processing executed as an evaluation unit, and the processing of S220 and S230 corresponds to an example of processing executed as a calculation unit.

[0067] 2. Other Embodiments Although the embodiments of the present disclosure have been described above, it goes without saying that the present disclosure is not limited to the above-described embodiments and can take on various forms.

[0068] (2a) In the above embodiment, the machine learning model 2 has, as the feature 41 corresponding to each of one or more expression elements 40, an explanatory variable representing the appearance time 41a of the corresponding expression element 40. However, the feature 41 related to the expression element 40 is not limited to the appearance time 41a.

[0069] For example, the machine learning model 2 may have, as the feature quantity 41 corresponding to each of one or more expression elements 40, an explanatory variable indicating whether or not the corresponding expression element 40 exists. Alternatively, the machine learning model 2 may have a category value related to the distribution of one or more expression elements 40 in a video advertisement as the feature 41 related to one or more expression elements 40. The category value may be a value that can identify at least one of the presence or absence, appearance duration, and appearance time of each of the one or more expression elements 40 in the video advertisement. The category value may be a value that represents a classification (i.e., a value that represents a class to which the video advertisement belongs) when the video advertisement is classified into indicators based on at least one of the presence or absence, appearance duration, and appearance time of each of the one or more expression elements 40.

[0070] (2b) In the above embodiment, feature quantities 41 related to one or more expression elements 40 included in the video advertisement are input to the machine learning model 2. However, the feature quantities input to the machine learning model 2 are not limited to this. For example, in addition to feature quantities 41 related to expression elements 40, feature quantities related to the product category in the video advertisement, feature quantities related to the appeal content of the video advertisement, etc. may also be input to the machine learning model 2.

[0071] (2c) In the above embodiment, the processor 11 identifies expression elements 40 with a relatively high contribution rate 52 from among the expression elements 40 related to the feature amount 41 input to the machine learning model 2. However, the processor 11 is not limited to the expression elements 40 with a relatively high contribution rate 52, and may identify expression elements 40 with a relatively low contribution rate 52. The processor 11 may identify at least one of the expression elements 40 with a relatively high contribution rate 52 and the expression elements 40 with a relatively low contribution rate 52.

[0072] 8, in addition to the top three expression elements 40 with the highest contribution rate 52, the bottom two expression elements 40 with the lowest contribution rate 52 are also displayed. That is, the processor 11 identifies both the expression elements 40 with relatively high contribution rates 52 and the expression elements 40 with relatively low contribution rates 52, and displays them on the display of the user interface 14.

[0073] (2d) In the above embodiment, the evaluation system 100 includes one machine learning model 2. When feature quantities 41 related to one or more expression elements 40 included in a video advertisement are input, the one machine learning model 2 is configured to output an evaluation value 51 calculated using one or more evaluation indicators 5 for the video advertisement based on the feature quantities 41, and a contribution degree 52 corresponding to each of the one or more expression elements 40.

[0074] However, one machine learning model 2 may be configured to output an evaluation value 51 calculated using one evaluation index 5 for the video advertisement. One machine learning model 2 may be configured to output contributions 52 corresponding to each of one or more expression elements 40 for the one evaluation index 5.

[0075] 9, the evaluation system 100 may include multiple machine learning models (e.g., a first machine learning model 2a and a second machine learning model 2b) corresponding to multiple evaluation indexes 5. In other words, the evaluation system 100 may include a machine learning model for each evaluation index 5 as one machine learning model 2.

[0076] The first machine learning model 2a as one machine learning model 2 may be configured to output an evaluation value 51 ("94.1" in FIG. 9) calculated using the first evaluation index 5a as one evaluation index 5. The first machine learning model 2a may be configured to output, for the first evaluation index 5a, a first contribution degree 52a as a contribution degree 52 corresponding to each of "expression element A" to "expression element D" as one or more expression elements 40.

[0077] Similarly, the second machine learning model 2b as one machine learning model 2 may be configured to output an evaluation value 51 ("73.5" in FIG. 9) calculated using the second evaluation index 5b as one evaluation index 5. The second machine learning model 2b may be configured to output, for the second evaluation index 5b, second contribution degrees 52b as contribution degrees 52 corresponding to each of "expression element E"-"expression element H" as one or more expression elements 40.

[0078] The second machine learning model 2b may be input with some or all of the one or more expression elements 40 ("expression element A" to "expression element D" in FIG. 9) input to the first machine learning model 2a. The second machine learning model 2b may calculate an evaluation value 51 using a second evaluation index 5b based on some or all of the one or more expression elements 40 input to the first machine learning model 2a. The second machine learning model 2b may output second contribution degrees 52b corresponding to some or all of the one or more expression elements 40 input to the first machine learning model 2a.

[0079] The machine learning model 2 may be configured to include at least one other machine learning model 2. For example, as shown in Fig. 9, the machine learning model 2 may be configured to include a first machine learning model 2a and a second machine learning model 2b.

[0080] (2e) Multiple functions possessed by one component in the above embodiments may be realized by multiple components, or one function possessed by one component may be realized by multiple components. Multiple functions possessed by multiple components may be realized by one component, or one function realized by multiple components may be realized by one component. Part of the configuration of the above embodiments may be omitted. At least part of the configuration of the above embodiments may be added to or substituted for the configuration of another of the above embodiments.

[0081] (2f) The present disclosure can be realized in various forms other than the evaluation device described above, such as a system including the evaluation device as a component, a computer program for causing a computer to function as the evaluation device, a non-transitory tangible recording medium such as a semiconductor memory on which the computer program is recorded, an evaluation method, etc.

[0082] [Technical idea disclosed in this specification] [Item 1] An evaluation device for evaluating a video advertisement, an acquisition unit configured to acquire features related to one or more expression elements included in the video advertisement; an evaluation unit configured to input the feature amounts into a trained machine learning model and acquire, from the machine learning model, an evaluation value for the video advertisement output by the machine learning model using one or more evaluation indexes based on the feature amounts; Equipped with the one or more representations include at least one of visual information and audio information; The one or more evaluation indicators include one or more indicators related to advertising effectiveness. Evaluation equipment.

[0083] [Item 2] The evaluation device according to item 1, the one or more expressions include auditory information; Evaluation equipment.

[0084] [Item 3] The evaluation device according to item 1, the one or more expressions include both visual and auditory information; Evaluation equipment.

[0085] [Item 4] The evaluation device according to any one of items 1 to 3, the acquisition unit is configured to acquire, as the feature amount, appearance times of each of the one or more expression elements in the video advertisement; the evaluation unit is configured to input the appearance times corresponding to each of the one or more expression elements as the feature amounts to the machine learning model. Evaluation equipment.

[0086] [Item 5] Item 4. The evaluation device according to item 4, The appearance time is a total value of the time that the corresponding expression element appears in the video advertisement for each of the one or more expression elements. Evaluation equipment.

[0087] [Item 6] The evaluation device according to any one of items 1 to 5, A calculation unit is further provided, the one or more expressions include a plurality of elements; The calculation unit is configured to calculate a contribution degree, which is a degree to which each of the plurality of elements contributes to each of the one or more evaluation indexes. Evaluation equipment.

[0088] [Item 7] Item 6. The evaluation device according to item 6, the calculation unit is configured to identify at least one element having a relatively high degree of contribution from among the plurality of elements; Evaluation equipment.

[0089] [Item 8] 1. A computer-implemented evaluation method for evaluating video advertisements, comprising: acquiring features relating to one or more expression elements included in the video advertisement; inputting the feature amounts into a trained machine learning model, and acquiring, from the machine learning model, an evaluation value for the video advertisement output by the machine learning model using one or more evaluation indexes based on the feature amounts; Including, the one or more representations include at least one of visual information and audio information; The one or more evaluation indicators include one or more indicators related to advertising effectiveness. Evaluation method.

[0090] [Item 9] Item 9. A computer program for causing a computer to execute the evaluation method according to Item 8. [Explanation of symbols]

[0091] 1...Evaluation device, 11...Processor, 2...Machine learning model, 40...Expression element, 41...Feature, 41a...Appearance time, 5, 5a, 5b...Evaluation index, 51...Evaluation value, ADG...Past advertisement group, IL...List of expression elements, EV...Evaluation data.

Claims

1. An evaluation device for evaluating a video advertisement, an acquisition unit configured to acquire features related to one or more expression elements included in the video advertisement; an evaluation unit configured to input the feature amounts into a trained machine learning model and acquire, from the machine learning model, an evaluation value for the video advertisement output by the machine learning model using one or more evaluation indexes based on the feature amounts; Equipped with the one or more representations include at least one of visual information and audio information; The one or more evaluation indexes are one or more indexes related to advertising effectiveness, including an index related to advertising expression effectiveness; Evaluation equipment.

2. An evaluation device according to claim 1, The indicators relating to the effectiveness of the advertisement expression include at least one of an attractiveness presentation level indicating the degree to which the attractiveness of the brand or product in the video advertisement is presented, and an attention arousal level indicating the degree to which attention is drawn to the video advertisement. Evaluation equipment.

3. The evaluation device according to claim 1 , the acquisition unit is configured to acquire, as the feature amount, appearance times of the one or more expression elements in the video advertisement; the evaluation unit is configured to input the appearance times corresponding to each of the one or more expression elements into the machine learning model as the feature amounts; Evaluation equipment.

4. The evaluation device according to claim 3, The appearance time is a total value of the time that the corresponding expression element appears in the video advertisement for each of the one or more expression elements. Evaluation equipment.

5. An evaluation device for evaluating a video advertisement, an acquisition unit configured to acquire features related to one or more expression elements included in the video advertisement; an evaluation unit configured to input the feature amounts into a trained machine learning model and acquire, from the machine learning model, an evaluation value for the video advertisement output by the machine learning model using one or more evaluation indexes based on the feature amounts; Equipped with the one or more representations include at least one of visual information and audio information; the one or more evaluation indicators include one or more indicators related to advertising effectiveness; the acquisition unit is configured to acquire, as the feature amount, appearance times of the one or more expression elements in the video advertisement; the appearance time of each of the one or more expression elements is represented by a start time and an end time or a start time and an elapsed time at which each of the one or more expression elements appears; the evaluation unit is configured to input the appearance times corresponding to each of the one or more expression elements into the machine learning model as the feature amounts; Evaluation equipment.

6. The evaluation device according to any one of claims 1 to 5, the one or more expressions include auditory information; Evaluation equipment.

7. The evaluation device according to any one of claims 1 to 5, the one or more expressions include both visual and auditory information; Evaluation equipment.

8. The evaluation device according to any one of claims 1 to 5, A calculation unit is further provided, the one or more expressions include a plurality of elements; the calculation unit is configured to calculate a contribution degree, which is a degree to which each of the plurality of elements contributes to each of the one or more evaluation indexes, based on the evaluation value; Evaluation equipment.

9. The evaluation device according to claim 8, the calculation unit is configured to identify at least one element having a relatively high degree of contribution from among the plurality of elements; Evaluation equipment.

10. 1. A computer-implemented evaluation method for evaluating video advertisements, comprising: acquiring features relating to one or more expression elements included in the video advertisement; inputting the feature amounts into a trained machine learning model, and acquiring, from the machine learning model, an evaluation value for the video advertisement output by the machine learning model using one or more evaluation indexes based on the feature amounts; Including, the one or more representations include at least one of visual information and audio information; The one or more evaluation indexes are one or more indexes related to advertising effectiveness, including an index related to advertising expression effectiveness; Evaluation method.

11. A computer program for causing a computer to execute the evaluation method according to claim 10.

12. 1. A computer-implemented evaluation method for evaluating video advertisements, comprising: acquiring features relating to one or more expression elements included in the video advertisement; inputting the feature amounts into a trained machine learning model, and acquiring, from the machine learning model, an evaluation value for the video advertisement output by the machine learning model using one or more evaluation indexes based on the feature amounts; Including, the one or more representations include at least one of visual information and audio information; the one or more evaluation indicators include one or more indicators related to advertising effectiveness; acquiring the feature amounts includes acquiring, as the feature amounts, appearance times of the one or more expression elements in the video advertisement; the appearance time of each of the one or more expression elements is represented by a start time and an end time or a start time and an elapsed time at which each of the one or more expression elements appears; acquiring the evaluation value from the machine learning model includes inputting the appearance times corresponding to each of the one or more expression elements into the machine learning model as the feature amounts; Evaluation method.

13. A computer program for causing a computer to execute the evaluation method according to claim 12.

Images