Examination system, examination method, program

Abstract

To provide an inspection system, an inspection method, and a program capable of properly and efficiently inspecting video advertisements. [Solution] The inspection system comprises an acquisition means 31 for acquiring information including a video advertisement, a determination means 34 for determining whether the video advertisement contains predetermined issues, and a provision means 35 for providing information regarding the predetermined issues when the video advertisement contains the predetermined issues. Also, the inspection method includes a computer executing the steps of acquiring information including the video advertisement, determining whether the video advertisement contains predetermined issues, and providing information regarding the predetermined issues when the video advertisement contains the predetermined issues.

Description

[Technical Field]

[0001] The present invention relates to an examination system, an examination method, and a program. [Background technology]

[0002] BACKGROUND ART Conventionally, before a video advertisement such as a television advertisement is placed (broadcast, distributed, etc.), the video advertisement is screened by a media (media provider) such as a television broadcasting station (for example, Non-Patent Document 1).

[0003] This review generally involves a business type review and a material review, and the material review examines whether the content of the advertising material violates broadcasting ethics and complies with laws, regulations, broadcasting standards, etc. [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Kokokusha Co., Ltd., "What is CM Review? Business Type Review and Material Review," [online], [Retrieved May 28, 2024], Internet<URL:https: / / cm.kokoku-direct.jp / column / Examination> Summary of the Invention [Problem to be solved by the invention]

[0005] However, because the results of material review depend on the experience and knowledge of the reviewer in charge of the review, bias may occur depending on the reviewer. This may make it difficult to conduct proper material review. Furthermore, because specialized knowledge of laws and regulations related to advertising materials is required for each type of video advertisement, it may be difficult for a single reviewer to efficiently review various types of video advertisement materials.

[0006] The present invention has been made in consideration of the above-mentioned problems, and aims to provide an inspection system, inspection method, and program that are capable of performing inspection processing of video advertisements appropriately and efficiently. [Means for solving the problem]

[0007] In order to solve the above problems, firstly, the present invention provides an inspection system comprising an acquisition means for acquiring information including a video advertisement, a determination means for determining whether or not the video advertisement contains specified criticisms, and a provision means for providing information regarding the specified criticisms when the video advertisement contains the specified criticisms.

[0008] According to this invention, when information including a video advertisement is acquired, it is determined whether the video advertisement contains predetermined issues, and if the video advertisement contains predetermined issues, information regarding the predetermined issues is provided. Therefore, for example, it is possible to provide predetermined issues as inspection results without relying on the experience or knowledge of the inspector, and without the inspector needing various specialized knowledge. This allows for proper and efficient inspection of video advertisements.

[0009] Secondly, the present invention provides an inspection method in which a computer executes the following steps: acquiring information including a video advertisement; determining whether the video advertisement includes specified criticisms; and, if the video advertisement includes the specified criticisms, providing information regarding the specified criticisms.

[0010] Thirdly, the present invention provides a program for enabling a computer to perform the following functions: acquire information including a video advertisement; determine whether the video advertisement contains specified indications; and, if the video advertisement contains the specified indications, provide information regarding the specified indications. [Effects of the Invention]

[0011] According to the inspection system, inspection method, and program of the present invention, inspection processing of video advertisements can be carried out properly and efficiently. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a diagram showing an outline of the basic configuration of an examination system according to one embodiment of the present invention; [Figure 2] FIG. 2 is a block diagram showing the configuration of an examination device. [Figure 3] This is a functional block diagram to explain the functions that play a major role in the examination system. [Figure 4] FIG. 10 is a diagram illustrating an example of the configuration of acquired data. [Figure 5] FIG. 2 is a diagram illustrating an example of the configuration of video advertisement data. [Figure 6] FIG. 10 is a diagram illustrating an example of the configuration of conversion data. [Figure 7] FIG. 10 is a diagram illustrating an example of the configuration of pointed out item data. [Figure 8] 1 is a flowchart showing an example of the main processing of an examination system according to one embodiment of the present invention. [Figure 9] FIG. 4 is a diagram illustrating an example of the configuration of first learning data. [Figure 10] FIG. 10 is a diagram illustrating an example of the configuration of second learning data. [Figure 11] FIG. 10 is a diagram illustrating another example of the configuration of the indications. [Figure 12] A diagram showing an example of the division of functions of the examination system between a terminal device and an examination device. DETAILED DESCRIPTION OF THE INVENTION

[0013] Hereinafter, an embodiment of the present invention will be described in detail with reference to the accompanying drawings. However, this embodiment is an example and the present invention is not limited to this embodiment.

[0014] (1) Basic configuration of the examination system Fig. 1 is a diagram showing a basic configuration of an examination system according to one embodiment of the present invention. As shown in Fig. 1, in the information processing system according to this embodiment, when a user (e.g., an advertisement creator or an examiner) inputs a video advertisement using a terminal device 10, information including the input video advertisement is transmitted to an examination device 20 via a communication network NW (network) such as the Internet or a LAN (Local Area Network).

[0015] In this embodiment, the inspection device 20 is configured to determine whether or not a video advertisement contains specified indications, and if the video advertisement contains specified indications, to provide the user with information regarding the specified indications.

[0016] The terminal device 10 may be, for example, a terminal device operated by an individual user (for example, a mobile terminal, a smartphone, a PDA, a personal computer, a television receiver with a two-way communication function (including a so-called multi-function smart television), etc.). The terminal device 10 may also be configured to be able to execute an application for instructing the examination device 20 to execute an examination process for the advertising video and displaying the results of the examination process on a display unit (not shown) such as an LCD (liquid crystal display) monitor.

[0017] The examination device 20 is configured to communicate with the terminal device 10 via a communication network NW. The examination device 20 may be a device operated by an individual user, such as a mobile terminal, a smartphone, a PDA, a personal computer, or a television receiver with a two-way communication function, or may be a server computer.

[0018] (2) Configuration of the testing device The configuration of the examination device 20 will be described with reference to Fig. 2. Fig. 2 is a block diagram showing the internal configuration of the examination device 20. As shown in Fig. 2, the examination device 20 includes a CPU (Central Processing Unit) 21, a ROM (Read Only Memory) 22, a RAM (Random Access Memory) 23, a storage device 24, a display processing unit 25, a display unit 26, an input unit 27, and a communication interface unit 28, and is provided with a bus 20a for transmitting control signals or data signals between the units.

[0019] When power is applied to the examination device 20, the CPU 21 loads various programs stored in the ROM 22 or the storage device 24 into the RAM 23 and executes them. In this embodiment, the CPU 21 reads and executes the programs stored in the ROM 22 or the storage device 24, thereby realizing the functions of an acquisition means 31, a detection means 32, a conversion means 33, a determination means 34, and a provision means 35 (shown in FIG. 3) described below.

[0020] The storage device 24 may be a non-volatile storage device such as a flash memory, an SSD (Solid State Drive), a magnetic storage device (e.g., an HDD (Hard Disk Drive), a floppy disk (registered trademark), a magnetic tape, or the like), an optical disk, or a volatile storage device such as a RAM, and stores programs executed by the CPU 21 and data referenced by the CPU 21. The storage device 24 also stores acquired data (shown in FIG. 4), converted data (shown in FIG. 6), and indicated item data (shown in FIG. 7), which will be described later.

[0021] The display processing unit 25 displays the display data provided by the CPU 21 on the display unit 26. The display unit 26 is, for example, an LCD monitor including thin film transistors arranged in a matrix on a pixel-by-pixel basis, and displays the data to be displayed on the display screen by driving the thin film transistors based on the display data.

[0022] When the examination device 20 is a button input type device, the input unit 27 has a button group including a plurality of instruction input buttons such as direction buttons and decision buttons for accepting user operation input, and a button group including a plurality of instruction input buttons such as a numeric keypad, and includes an interface circuit for recognizing the pressing (operation) input of each button and outputting it to the CPU 21.

[0023] If the examination device 20 is a touch panel input device, the input unit 27 mainly accepts touch panel input by touching the display screen with a fingertip or a pen. The touch panel input method may be a known method such as a capacitance method.

[0024] Furthermore, if the examination device 20 is a device capable of voice input, the input unit 27 may be configured to include a microphone for voice input, or may be equipped with an interface circuit for outputting voice data input via an external microphone to the CPU 21. Furthermore, if the examination device 20 is a device capable of inputting moving images and / or still images, the input unit 27 may be configured to include a digital camera or digital video camera for image input, or may be equipped with an interface circuit for receiving image data captured by an external digital camera or digital video camera and outputting it to the CPU 21.

[0025] The communication interface unit 28 includes an interface circuit for communicating with other devices (for example, the terminal device 10, etc.) via the communication network NW.

[0026] (3) Overview of each function in the examination system The functions realized by the examination system of this embodiment will be described with reference to Fig. 3. Fig. 3 is a functional block diagram for explaining the functions that play a major role in the examination system of this embodiment. In the functional block diagram of Fig. 3, an acquisition means 31, a determination means 34, and a provision means 35 correspond to the main components of the examination system of the present invention. The other means (detection means 32 and conversion means 33) are not necessarily essential components, but are components that make the present invention even more preferable.

[0027] The acquisition means 31 has a function of acquiring information including video advertisements.

[0028] The function of the acquisition means 31 is realized, for example, as follows. For example, when an application for instructing the execution of an examination process is running on the terminal device 10 and the user specifies a video advertisement to be examined on the application, the terminal device 10 transmits data of the specified video advertisement (video advertisement data) to the examination device 20 via the communication network NW. Here, the video advertisement data may be associated with identification information for identifying the video advertisement (e.g., a video advertisement ID or a file name of the video advertisement), an advertisement category that is a category of the goods or services to be advertised (e.g., "quasi-drugs," "foods," "health foods," "drinks," etc.), and an advertisement type that indicates the type of video advertisement (e.g., "commercial," "mail-order program," etc.). In addition, the advertisement category and advertisement type corresponding to the video advertisement may be specified by the user when the video advertisement to be examined is specified on the terminal device 10.

[0029] Meanwhile, when the CPU 21 of the examination device 20 receives (acquires) information including a video advertisement (including a video advertisement ID, advertisement category, and advertisement type) transmitted from the terminal device 10 via the communication interface unit 28, the CPU 21 stores the information including the video advertisement in association with the acquisition date and time (reception date and time) of the information in the acquired data shown in FIG. 4, for example. The acquired data is data in which the video advertisement ID, advertisement category, advertisement type, and video advertisement data of the video advertisement are described in association with each other for each acquisition date and time of the information including the video advertisement. In this way, the CPU 21 can acquire information including the video advertisement.

[0030] The detection means 32 has a function of detecting a predetermined object from a video advertisement. Here, the object may be anything that can be subject to examination, such as text data (e.g., a script (dialogue) or a character string), numbers, symbols, sounds, figures (e.g., a graph, a company logo, a product logo, a service logo), images (video and / or still images), characters appearing in the video advertisement, or objects or backgrounds appearing in the video advertisement.

[0031] Furthermore, the detection means 32 may detect a predetermined object using machine learning, which makes it possible to easily detect one or more objects included in a video advertisement using machine learning.

[0032] The function of the detection means 32 is realized, for example, as follows. When the CPU 21 of the examination device 20 acquires information including a video advertisement based on the function of the acquisition means 31 described above, it detects one or more objects included in the acquired video advertisement. Here, the CPU 21 may detect, as an object, one or more pieces of text data (script (dialogue), character strings, etc.), numbers, symbols, etc. included in the video advertisement, for example, using OCR (Optical Character Recognition / Reader) technology based on machine learning. The CPU 21 may also detect, as an object, one or more figures, images, characters, objects, backgrounds, etc. included in the video advertisement, for example, using image recognition technology based on machine learning (for example, object recognition, object detection, etc.). Furthermore, the CPU 21 may detect, as an object, the text data of the script (dialogue) included in the video advertisement, by converting audio data (for example, script (dialogue)) included in the video advertisement into text data using well-known voice recognition technology.

[0033] Furthermore, for each of the detected one or more objects, the CPU 21 may detect a presentation mode (e.g., presentation content, presentation time, presentation size, presentation position, etc.) of the object in the video advertisement. Here, the presentation time of the object in the video advertisement may be detected using, for example, a timer. Furthermore, as shown in FIG. 5, the CPU 21 may store each of the detected one or more objects (in the example shown in the figure, "script (dialogue)," "company logo," "price display," etc.) in association with the presentation mode of the object for each of the video advertisement data stored in the acquired data. For example, as the presentation mode of an object called "script (dialogue)," the content of the "script (dialogue)" may be stored. Furthermore, as the presentation mode of an object called "company logo," the display time, display size, and / or display position of the "company logo" may be stored. Furthermore, as the presentation mode of an object called "price display," the display time, display size, and / or display position of the "price display" may be stored.

[0034] The conversion means 33 has a function of converting text data included in the video advertisement into vector data, which is data obtained by vectorizing the text data.

[0035] The function of the conversion means 33 is realized, for example, as follows. For example, the CPU 21 of the examination device 20 detects text data (for example, character strings detected by OCR technology or text data converted from audio data) included in a video advertisement based on the function of the detection means 32. The CPU 21 then converts the text data into vector data by vectorizing the text data using a well-known vectorization technology. Here, the vectorization technology may be, for example, the Embeddings API provided by OpenAI, Inc., a U.S. company, or another vectorization technology. After converting the text data into vector data, the CPU 21 stores the converted vector data in, for example, the conversion data shown in FIG. 6. The conversion data is data in which, for each acquisition date and time of the video advertisement, identification information of the video advertisement (in the example shown in the figure, a video advertisement ID), text data in the video advertisement data, and vector data of the text data are described in association with each other. In this way, the CPU 21 can convert the text data included in the video advertisement into vector data.

[0036] The determining means 34 has a function of determining whether or not a predetermined indication is included in the video advertisement.

[0037] Here, the predetermined issue may include the inclusion of a predetermined object in the video advertisement. This makes it possible to determine that the video advertisement contains an issue of concern when, for example, an inappropriate object (predetermined object) is included in the video advertisement. Furthermore, the predetermined object may include multiple expressions that appear in a predetermined order (for example, first a negative expression (such as a line or an image), then an expression (such as a line or an image) that identifies the advertised product or service, such as "Discover this product," and then a positive expression (such as a line or an image)). This makes it possible to determine whether the structure, story, etc. of the video advertisement fall under the issue of concern.

[0038] Furthermore, the predetermined indication may include a fact that a predetermined object is not included in the video advertisement. This makes it possible to determine that the video advertisement contains an indication when, for example, an object (predetermined object) that should be included in the video advertisement is not included in the video advertisement.

[0039] Furthermore, the determination means 34 may determine that the video advertisement includes a predetermined indication when the presentation manner of the detected object in the video advertisement satisfies a predetermined condition. This makes it possible to determine whether the video advertisement includes a predetermined indication based on the presentation manner of the object in the video advertisement.

[0040] The predetermined condition may also include whether the presentation time of the detected object in the video advertisement is equal to or longer than a predetermined time, thereby making it possible to determine whether the video advertisement includes a predetermined indication based on the presentation time of the object in the video advertisement.

[0041] The function of the determination means 34 is realized, for example, as follows. The CPU 21 of the inspection device 20 detects one or more objects in a video advertisement based on the function of the detection means 32 described above, and converts the text data in the video advertisement into vector data based on the function of the conversion means 33 described above, and then accesses the findings data shown in FIG. 7, for example. The findings data is data that describes, for example, for each video advertisement ID of multiple video advertisements that have been inspected in advance, an advertisement category, an advertisement type, one or more objects in the video advertisement and the presentation manner of the objects, and findings that are findings regarding the objects in the inspection process of the video advertisement, all in association with each other. Here, multiple different findings may be associated with the same object and the presentation manner of the object.

[0042] In the inspection process (material inspection), various issues may be raised about objects included in the video advertisement. Examples of issues that may be raised in the inspection process are shown below. - If a video ad does not include necessary objects (for example, the title of the video ad, an indication that the video ad is END, an indication that it is a TV shopping program, a credit indication for the provider, a return policy indication, a personal information protection policy indication, an indication of "quasi-drugs" when the ad category is "quasi-drugs", an indication of the country of origin when the ad category is "miscellaneous goods", etc.), this will be pointed out. - If a video ad contains inappropriate content (for example, discriminatory language, false or exaggerated language, unfounded language, loud or subliminal language, language that violates the Pharmaceutical Affairs Act or Health Promotion Act when the ad category is "drugs" or "quasi-drugs," language that is offensive to viewers, etc.), we will point it out. - If a video ad contains an object that lacks sufficient content (for example, a graph without a research summary or source, an image without a search box, or an image that does not comply with the material credit rules), we will point it out. - If a video ad contains an object that violates the display time limit (for example, a phone number, an END message, a QR code (registered trademark) message, a program title, a TV shopping message, a sponsor credit message, etc.), we will point it out. - We will point out any inappropriate objects in the structure or story of a video ad (for example, first showing negative language (lines, images, etc.), then showing language (lines, images, etc.) that identifies the advertised product or service, such as "discover this product," followed by positive language (lines, images, etc.)). - We will point out any inappropriate objects (for example, experts such as doctors or university professors explaining experimental examples or clinical results) that are included as components of video advertisements.

[0043] The contents of the pointed out items data may be added to and / or updated by the CPU 21 based on information input using the input unit 27, or may be added to and / or updated by the CPU 21 based on information input from outside the examination device 20 via the communication interface unit 28, for example.

[0044] For each of the one or more detected objects, the CPU 21 determines whether or not there is a plurality of indications in the indication data that are associated with the advertisement category and advertisement type of the acquired video advertisement, and an object whose presentation manner satisfies a predetermined condition similar to that of the object of the acquired video advertisement. Furthermore, if there is one or more indications corresponding to any of the one or more detected objects, the CPU 21 may determine that the acquired video advertisement includes an indication and extract (acquire) the indication. Here, the predetermined condition may include, for example, whether the presentation time of the object is equal to or longer than a predetermined time, whether the presentation size of the object is equal to or shorter than a predetermined value, or whether the distance between the presentation position of the object and a predetermined position is equal to or shorter than a predetermined value. Furthermore, the predetermined condition may include whether a predetermined similarity condition is satisfied between vector data of at least a portion of text data included in the acquired video advertisement and vector data in the indication data (including vector data of the portion indicated by the indication). Here, the predetermined similarity condition may include, for example, that the cosine similarity between at least a part of vector data of text data included in the acquired video advertisement and the vector data in the indication data is equal to or greater than a predetermined value (e.g., 0.9). In this way, the CPU 21 can determine whether the video advertisement includes a predetermined indication.

[0045] The providing means 35 has a function of providing information on a predetermined indication when the video advertisement includes the predetermined indication.

[0046] The function of the providing means 35 is realized, for example, as follows. When the CPU 21 of the examination device 20 determines, based on the function of the determining means 34 described above, that the video advertisement includes one or more points of criticism, it may transmit information about each of the one or more points of criticism to the terminal device 10 via the communication interface unit 28 and the communication network NW. Then, the terminal device 10 may display the information about each of the one or more points of criticism received from the examination device 20 on a display unit (not shown). In this way, the CPU 21 can provide the user of the terminal device 10 with information about each of the one or more points of criticism included in the video advertisement.

[0047] The information on the pointed out matters may be composed of text data or image data, etc. Furthermore, when the information on the pointed out matters is composed of audio data, the terminal device 10 may output the information on the pointed out matters from an audio output device such as a speaker.

[0048] (4) Main processing flow of the examination system of this embodiment Next, an example of the flow of the main processing performed by the examination system of this embodiment will be described with reference to the flowchart of FIG.

[0049] First, the CPU 21 of the examination device 20 acquires information including a video advertisement based on the function of the acquisition means 31 (step S100). Here, the CPU 21 of the examination device 20 may detect a predetermined object from the video advertisement based on the function of the detection means 32. The CPU 21 may also detect the predetermined object using machine learning based on the function of the detection means 32. Furthermore, the CPU 21 may convert text data included in the video advertisement into vector data, which is data obtained by vectorizing the text data, based on the function of the conversion means 33.

[0050] Next, the CPU 21 of the examination device 20 determines whether or not the video advertisement includes a predetermined item of indication based on the function of the determination means 34 (step S102). Here, the CPU 21 of the examination device 20 may determine that the video advertisement includes a predetermined item of indication when the presentation mode in the video advertisement of the object detected based on the function of the detection means 32 satisfies a predetermined condition.

[0051] Next, the CPU 21 of the inspection device 20 provides information on the specified pointed out matter when the video advertisement includes the specified pointed out matter based on the function of the providing means 35 (step S104).

[0052] As described above, according to the inspection system, inspection method, and program of this embodiment, when information including a video advertisement is acquired, it is determined whether the video advertisement contains certain specified issues, and if the video advertisement contains certain specified issues, information regarding the certain specified issues is provided. This makes it possible to provide certain specified issues as inspection results without relying on the experience or knowledge of the inspector, and without the inspector needing various specialized knowledge. This allows for appropriate and efficient inspection of video advertisements.

[0053] The program of the present invention may be stored in a computer-readable storage medium. The storage medium on which this program is recorded may be the ROM 22, RAM 23, or storage device 24 of the examination device 20 shown in Figure 2. The storage medium may also be a CD-ROM or the like that can be read by being inserted into a program reading device such as a CD-ROM drive. Furthermore, the storage medium may be a magnetic tape, cassette tape, flexible disk, MO / MD / DVD, or semiconductor memory.

[0054] The above-described embodiments have been described to facilitate understanding of the present invention, and are not intended to limit the present invention. Therefore, each element disclosed in the above embodiments is intended to include all design modifications and equivalents that fall within the technical scope of the present invention.

[0055] Modifications of the above-described embodiment will now be described. (Variation 1) In the above embodiment, the determination means 34 determines whether a video advertisement contains a specific indication by determining whether there is an indication associated with the advertisement category, advertisement type, and object whose presentation style satisfies a specific condition similar to that of the object in the acquired video advertisement, among the multiple indications in the indication data. However, the present invention is not limited to this case. For example, the determination means 34 may determine whether a video advertisement contains a specific indication based on the video advertisement and a trained model based on machine learning using a previously acquired video advertisement as first training data. In this case, using the trained model makes it easy to determine whether a video advertisement contains a specific indication.

[0056] The function of the determination means 34 in this modified example is realized, for example, as follows: First, the CPU 21 of the inspection device 20 acquires information including the video advertisement (including the advertisement category and advertisement type) based on the function of the acquisition means 31. The CPU 21 may also detect one or more objects in the video advertisement and the presentation manner of the objects based on the function of the detection means 32. Then, the CPU 21 may input the advertisement category and advertisement type of the video advertisement and the detected one or more objects and the presentation manner of the objects into a trained model based on machine learning using the first training data, thereby determining whether the video advertisement contains a predetermined indication.

[0057] An example of the first learning data is shown in FIG. 9. The first learning data shown in FIG. 9 is data in which, for each of a plurality of video advertisements acquired in advance, an advertisement category, an advertisement type, one or more objects in the video advertisement and the presentation manner of the objects, and a problem (correct label) that is a problem that is pointed out for the object in the inspection process of the video advertisement are described in association with each other. In this modification, when text data (a script (dialogue), a character string, etc.) is included in the video advertisement, the text data is stored as an object in the first learning data. Furthermore, the first learning data may associate multiple different problems (correct labels) with the same combination of advertisement category, advertisement type, object, and presentation manner of the object. As a result of machine learning, a trained model is constructed that indicates the relationship between the advertisement category, advertisement type, one or more objects and the presentation manner of the objects of the video advertisement, and the problem (correct label) that is pointed out for one or more objects in the video advertisement.

[0058] In addition, CPU21 may learn a model used to determine whether or not a specified indication is included in the acquired video advertisement based on the advertisement category, advertisement type, one or more objects, and the presentation manner of the objects, by machine learning using the first learning data.

[0059] In this case, for example, when a predetermined model learning instruction is input using the input unit 27, the CPU 21 may perform model learning using the first learning data shown in FIG. 9. The CPU 21 may perform learning using, for example, a time-series adaptive neural network model. Here, as the time-series adaptive neural network, for example, a recurrent neural network (RNN) or a long short-term memory (LSTM), which is an advanced version of an RNN, may be applied. Furthermore, the CPU 21 may perform learning using any one of a plurality of models, for example, a graph neural network (GNN) model, a convolutional neural network (CNN) model, a support vector machine (SVM) model, a fully connected neural network (FNN) model, a gradient boosting (HGB) model, a wave net (WN) model, and an extremely randomized trees (ExtraTrees) model. Furthermore, the CPU 21 may perform learning using any of a graph convolutional neural network (GCN) model, a graph attention network (GAT) model, and a graph convolutional LSTM (GC-LSTM) model, which are derivatives of GNN.

[0060] The trained model used to determine whether or not a video advertisement contains a predetermined indication may be provided in a device other than the inspection device 20. In this case, the CPU 21 may input the advertisement category and advertisement type of the acquired video advertisement, at least a part of the script in the video advertisement, and one or more objects and presentation modes of the objects in the video advertisement into the trained model provided in the other device, and receive (acquire) information related to the result of the determination made by the trained model (i.e., whether or not the video advertisement contains a predetermined indication) from the other device.

[0061] (Variation 2) In the above-described first modification, the determination means 34 determines whether a predetermined indication is included in an acquired video advertisement based on a trained model based on machine learning using first learning data in which text data (such as a script (dialogue) or character strings) is stored as an object when the video advertisement contains the text data. However, the present invention is not limited to this case. For example, the determination means 34 may determine whether a predetermined indication is included in an acquired video advertisement based on a trained model based on machine learning using second learning data in which vector data of the text data (such as a script (dialogue) or character strings) is stored as an object. Even in this case, the use of the trained model makes it possible to easily determine whether a predetermined indication is included in a video advertisement.

[0062] The function of the determination means 34 in this modified example is realized, for example, as follows. In this modified example, when text data is detected based on the function of the detection means 32, the CPU 21 of the examination device 20 may convert the text data into vector data based on the function of the conversion means 33. Then, the CPU 21 may input the advertisement category and advertisement type of the acquired video advertisement, and the detected one or more objects and the presentation manner of the objects, into a trained model based on machine learning using the second training data, thereby determining whether the acquired video advertisement contains a predetermined indication.

[0063] An example of the second learning data is shown in FIG. 10. The second learning data shown in FIG. 10 is data in which, for each of a plurality of video advertisements acquired in advance, the advertising category, the advertising type, one or more objects in the video advertisement and the presentation manner of the objects, and the issues (correct labels) that are issues that are pointed out for the objects in the video advertisement inspection process are described in association with each other. In this modification, when text data (such as a script (dialogue) or character strings) is included in the video advertisement, vector data converted from the text data is stored as an object in the second learning data. Furthermore, the second learning data may associate multiple different issues (correct labels) with the same combination of advertising category, advertising type, object, and presentation manner of the object. As a result of machine learning, a trained model is constructed that shows the relationship between the advertising category, advertising type, one or more objects, the presentation manner of the objects, and the issues that are pointed out for one or more objects in the video advertisement.

[0064] In addition, CPU21 may learn a model used to determine whether or not a specified indication is included in the acquired video advertisement based on the advertisement category, advertisement type, one or more objects, and the presentation manner of the objects of the acquired video advertisement, by machine learning using the second learning data.

[0065] (Variation 3) In the above embodiment, a case has been described as an example in which, when text data (a script (lines), a character string, etc.) is included in a video advertisement, vector data converted from the text data is stored in the indication data. However, the present invention is not limited to this case. For example, when text data is included in a video advertisement, the text data may be stored in the indication data without being converted into vector data, as shown in FIG. 11. In this case, it is possible to achieve the same effects as in the above embodiment.

[0066] The function of the determination means 34 in this modified example is realized, for example, as follows. When one or more pieces of text data are detected as objects from an acquired video advertisement, the CPU 21 of the inspection device 20 may determine whether or not there are any observations among the multiple observations in the observation data that correspond to the advertising category and advertisement type of the acquired video advertisement and an object that includes a portion that matches the text data in the video advertisement (a portion pointed out in the observations). Furthermore, when there are one or more observations that correspond to any of the one or more detected objects, the CPU 21 may determine that the acquired video advertisement includes observations and extract (acquire) the observations.

[0067] In addition, in the above-described embodiment and modified examples 1 to 3, the case where one examination device 20 is provided has been described as an example, but this is not limited to this. For example, multiple examination devices 20 may be provided, and in this case, the operation content and processing results, etc. on any of the examination devices 20 may be presented in real time on the other examination devices 20, or the processing results, etc. on any of the examination devices 20 may be shared among the multiple examination devices 20.

[0068] Furthermore, in the above-described embodiment and modifications 1 to 3, the examination device 20 is configured to realize the functions of the acquisition means 31, detection means 32, conversion means 33, discrimination means 34, and provision means 35, but this configuration is not limited to this. For example, another device (e.g., a management server that manages multiple examination devices 20) composed of a computer (e.g., a general-purpose personal computer or server computer) communicably connected to the examination device 20 via a communication network such as the Internet or a LAN may be provided. In this case, the examination device 20 and the other device can have substantially the same hardware configuration, so that the function of at least one of the means 31 to 35 described in the above-described embodiment and modifications 1 to 3 can be realized by the other device.

[0069] Furthermore, the function of at least one of the above means 31 to 35 may be realized by the terminal device 10 as shown in FIGS. 12(a) and 12(b). [Industrial Applicability]

[0070] The inspection system, inspection method, and program of the present invention as described above can perform inspection processing of video advertisements appropriately and efficiently, and can be suitably used, for example, in advertising production services and advertising media services, etc., and therefore has extremely great industrial applicability. [Explanation of symbols]

[0071] 10...Terminal device 20...Examination device 31…Acquisition means 32...Detection means 33...Conversion method 34...Discrimination means 35…Providing means

Claims

1. An acquisition means for acquiring information including a video advertisement; A determination means for determining whether or not the video advertisement includes a predetermined indication; providing means for providing information about the specified indications when the specified indications are included in the video advertisement; Examination system.

2. The predetermined indication includes that a predetermined object is included in the video advertisement. The examination system according to claim 1.

3. the predetermined object includes multiple representations that appear in a predetermined order; The examination system according to claim 2.

4. The predetermined indication includes that a predetermined object is not included in the video advertisement. The examination system according to claim 1.

5. a detection means for detecting a predetermined object from the video advertisement; The determination means determines that the video advertisement includes the specified indication when a presentation mode of the detected object in the video advertisement satisfies a specified condition. The examination system according to claim 1.

6. the detection means detects the predetermined object using machine learning; The examination system according to claim 5.

7. the predetermined condition includes a presentation time of the detected object in the video advertisement being equal to or longer than a predetermined time, or shorter than a predetermined time. The examination system according to claim 5.

8. The determination means determines whether the video advertisement includes the predetermined indication based on the video advertisement and a trained model based on machine learning using the video advertisement acquired in advance as training data. The examination system according to claim 1.

9. a conversion means for converting text data included in the video advertisement into vector data, the vectorized text data being vectorized data; The determination means determines whether the video advertisement includes the predetermined indication based on the converted vector data and a trained model based on machine learning using vector data of the language included in the video advertisement acquired in advance as training data. The examination system according to claim 1.

10. The computer obtaining information including a video advertisement; determining whether the video advertisement includes a predetermined indication; If the video advertisement includes the predetermined indication, providing information about the predetermined indication; Perform each step of Examination method.

11. On the computer, The ability to obtain information including video ads; a function of determining whether or not the video advertisement includes a predetermined indication; a function of providing information about the specified indications when the specified indications are included in the video advertisement; A program to achieve this.

Images