Content Rating System

Abstract

To provide a content evaluation system capable of evaluating a content material itself from multiple aspects. [Solution] The content evaluation system is a system that evaluates content material based on the emotions evoked in a recipient by the content material, and includes a content acquisition unit (10), a survey data acquisition unit (20), an evaluation parameter generation unit (30), and a content evaluation unit (40).

Description

[Technical field]

[0001] The present invention relates to a content rating system for rating content material based on emotions evoked in a recipient by the content material. [Background technology]

[0002] As described in Patent Document 1 below, the inventors of the present application are developing a system that can select a creator who is suitable for the emotions that the recipient wants to evoke through content, based on the emotions evoked in the recipient by the content material produced by the creator and the evaluation value of that emotions. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 7484000 Summary of the Invention [Problem to be solved by the invention]

[0004] Here, the inventors of the present application developed a method for evaluating the content material itself, and through extensive research, they came to the realization that even when content evaluation is referred to as a positive element (or a negative element), a wide variety of positive elements (or negative elements) are intertwined.

[0005] In view of the above circumstances, an object of the present invention is to provide a content evaluation system capable of evaluating a content material itself from multiple perspectives. [Means for solving the problem]

[0006] A content evaluation system according to a first aspect of the present invention is a content evaluation system for evaluating a content material based on an emotion evoked in a receiver by the content material, an evaluation parameter generation unit that generates emotion evaluation indices by visualizing emotions evoked in a receiver by the content material using an emotion visualization technique, the emotion labels being the emotions evoked in the receiver, and emotion scores by scoring the emotion labels; a content evaluation unit that evaluates the content material based on the emotion evaluation index generated by the evaluation parameter generation unit; Equipped with 、 The content evaluation unit is When dividing the emotion labels generated by the evaluation parameter generation unit into positive elements and negative elements and evaluating the content material from the emotion scores of each element included in the positive elements and the emotion scores of each element included in the negative elements, The content material is evaluated using an impression index based on the emotion label of an element having a higher emotion score from among the emotion scores of each element included in the positive element and the emotion scores of each element included in the negative element. .

[0007] According to the content evaluation system of the first invention, the content material itself can be evaluated from multiple perspectives based on the emotion evaluation index generated by the evaluation parameter generating section.

[0008] Specifically, consumers' latent needs, such as "I want to see this kind of content," cannot be fully captured by information such as the genre and content of the content material. However, by assigning emotional labels and emotional scores to the content material, the characteristics of the content material that evoke certain emotions can be extracted, making it possible to provide content that matches the viewer's emotions.

[0010] Also, No. 1 According to the content evaluation system of the invention, emotion labels, which are emotion evaluation indexes that visualize the emotions evoked in the recipient by the content material using emotion visualization technology, are divided into positive elements and negative elements, and the content is evaluated based on the emotion scores of each element included in the positive elements and the emotion scores of each element included in the negative elements.

[0011] Therefore, the content itself can be evaluated comprehensively by unraveling the intertwining of a wide variety of positive and negative elements, such as each element and its emotional score contained in the positive aspects, and each element and its emotional score contained in the negative aspects.

[0014] In addition, No. 1 Invention content evaluation system Leave The emotion scores of each element included in the positive elements are added as +, and the emotion scores of each element included in the negative elements are added as - to calculate the total emotion score. You may do so.

[0015] The total score calculated in this way for the entire content material allows for a comprehensive evaluation of the content material by unraveling the intertwining of the many and varied positive and negative factors.

[0016] Here, the total score can be calculated based on the entire video, an accumulation per second, or a partial item of the emotion score or a score at a partial point in time. In other words, the total score can be calculated as a score of 1 or more.

[0019] moreover, No. 1 According to the content evaluation system of the invention, since the elements with high emotional scores can be factors that can determine the impression of the content material, an impression index is generated that is determined through impression judgment based on the emotional labels of the elements with high emotional scores.

[0020] The impression index of the content material calculated in this manner can be used to comprehensively evaluate the content material by unraveling the intertwining of a wide variety of positive and negative elements.

[0021] Thus, the 1 According to the content evaluation system of the invention, the content material itself can be comprehensively evaluated from multiple angles.

[0022] No. 2 The content evaluation system of the present invention is 1 In the invention, The content evaluation unit classifies the content materials using the impression index determined by the content evaluation unit.

[0023] No. 2 According to the content evaluation system of the invention, the content material can be classified based on the impression index by the content evaluation unit, and the content material itself can be comprehensively evaluated from multiple angles.

[0024] No. 3 The content rating system of the invention comprises: A content evaluation system for evaluating content material based on emotions evoked by the content material in a recipient, comprising: an evaluation parameter generation unit that generates emotion evaluation indices by visualizing emotions evoked in a receiver by the content material using an emotion visualization technique, the emotion labels being the emotions evoked in the receiver, and emotion scores by scoring the emotion labels; a content evaluation unit that evaluates the content material based on the emotion evaluation index generated by the evaluation parameter generation unit; Equipped with The content evaluation unit determines, from a time transition of the emotion score, the emotion labels and The time when the change was significant is extracted as an emotional change starting point, and the content material is evaluated based on the emotional change starting point.

[0025] No. 3 According to the content evaluation system of the invention, since a change in emotion score can be the starting point of an impression of the content material, the emotion level and the point in time at which such an emotion score changes significantly are extracted as the starting point of the emotion change.

[0026] The emotional change starting points extracted in this way can be used, for example, to comprehensively evaluate content material by unraveling the intertwining of a wide variety of positive and negative elements, and can be used, for example, to insert advertisements (commercials) into content material.

[0027] Specifically, in AVOD (Advertising Video On Demand), advertisements are sometimes inserted at times unintended by the content creator, which can degrade the quality of the viewing experience. However, by assigning emotion scores to the content material and advertisements (advertisements as content material), it is possible to insert advertisements at the appropriate time.

[0028] Thus, the 3According to the content evaluation system of the invention, the content material itself can be comprehensively evaluated from multiple angles, and the results can be utilized.

[0029] No. 4 The content rating system of the invention comprises: A content evaluation system for evaluating content material based on emotions evoked by the content material in a recipient, comprising: an evaluation parameter generation unit that generates emotion evaluation indices by visualizing emotions evoked in a receiver by the content material using an emotion visualization technique, the emotion labels being the emotions evoked in the receiver, and emotion scores by scoring the emotion labels; a content evaluation unit that evaluates the content material based on the emotion evaluation index generated by the evaluation parameter generation unit; Equipped with The content evaluation unit is characterized in that it extracts, from a time progression of the emotion score, a portion of the content material corresponding to the emotion score portion exceeding a specific threshold, together with the emotion label corresponding to the emotion score.

[0030] No. 4 According to the content evaluation system of the invention, the portion of the emotion score that exceeds a specific threshold based on the change in emotion score can be a factor that determines the impact, i.e., the impression, of the content material. Therefore, the emotion score portion (time) where the emotion score exceeds a specific threshold and its emotion label are extracted.

[0031] The emotional score portion (time) where the emotional score exceeds a certain threshold and its emotional label extracted in this way can be used, for example, to comprehensively evaluate the content material by unraveling the intertwining of a wide variety of positive and negative elements, and can be used, for example, for creating thumbnails.

[0032] Here, the extraction of the portion exceeding the threshold value may include the portion before and after the portion exceeding the threshold value, and the extracted representative value may be used instead of the threshold value.

[0033] Thus, the 4 According to the content evaluation system of the invention, the content material itself can be comprehensively evaluated from multiple angles, and the results can be utilized.

[0034] No. 5 The content evaluation system of the present invention is 4 In the invention, The content evaluation unit extracts portions of the content material corresponding to the emotion score portions exceeding the particular threshold, and links the extracted emotion score portions of the content material together.

[0035] No. 5 According to the content evaluation system of the invention, an emotional score portion that exceeds a specific threshold can be a factor that can determine the impact, i.e., the impression, of the content material. Therefore, by linking the portions (time) of the content material that correspond to the extracted emotional score portions that exceed a specific threshold, it is possible to shorten the length of the content efficiently and accurately.

[0036] Thus, the 5 According to the content evaluation system of the invention, the content material itself can be comprehensively evaluated from multiple angles, and the results can be utilized.

[0037] No. 6 The content evaluation system of the present invention includes the first to second 5 In any of the inventions, The content evaluation unit provides the evaluation result to an advertisement transaction system.

[0038] No. 6 According to the content evaluation system of the invention, the evaluation results obtained by comprehensively evaluating the content material itself from multiple perspectives can be utilized as an index for advertising transactions. [Brief description of the drawings]

[0039] [Figure 1] 1 is a block diagram showing an overall configuration of a content evaluation system according to an embodiment of the present invention. [Diagram 2] 2 is an explanatory diagram showing the processing contents in a content evaluation unit of the content evaluation system in FIG. 1; [Diagram 3] 2 is an explanatory diagram showing the processing contents in a content evaluation unit of the content evaluation system in FIG. 1; [Figure 4]2 is an explanatory diagram showing the processing contents in a content evaluation unit of the content evaluation system in FIG. 1; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0040] The configuration of a content evaluation system according to an embodiment of the present invention will be described below with reference to FIG.

[0041] As shown in FIG. 1, the content evaluation system is a system that evaluates content material based on the emotions that the content material evokes in a recipient, and includes a content acquisition unit 10, a survey data acquisition unit 20, an evaluation parameter generation unit 30, and a content evaluation unit 40.

[0042] The content acquisition unit 10 acquires content material to be evaluated. The content material may be acquired from a content database DB1, an external server, or the like.

[0043] The survey data acquisition unit 20 is a means for acquiring information related to the content material and information related to the evaluation of the content material (for example, viewing data for broadcasts), and acquires the information from the survey database DB2 as well as from an external server.

[0044] Here, content material is, for example, commercial material data, which includes, in addition to the video data itself, commercial attribute information such as the commercial title, advertiser, production year, product, performers (talent / character), advertising agency name, producer name, 10-digit commercial code, background music (music / sound effects), narration, subtitles, language used, length of time, subtitles, format, distribution (broadcast) destination, and region.

[0045] The evaluation parameter generation unit 30 generates an emotion evaluation index by visualizing the emotion evoked in the receiver by the content material using emotion visualization technology, and generates an emotion label, which is the emotion evoked in the receiver, and an emotion score by scoring the emotion label.

[0046] Specifically, the evaluation parameter generation unit 30 uses a huge amount of content material, the objects depicted in the content material, the objects of expression (including expression forms that are not necessarily physical, such as background music, sound effects, natural sounds, performances, and utterances), and survey data on them as teacher data, and based on this, constructs an engine that generates an emotional evaluation index for the content material (constructs an engine that parameterizes the evaluation index).

[0047] The content evaluation unit 40 evaluates the content material based on the emotion evaluation index generated by the evaluation parameter generation unit 30.

[0048] The above is the configuration of the content evaluation system of this embodiment. In the above configuration, each of the processing units 10 to 40 is configured with hardware such as a CPU (Central Processing Unit), a ROM (Read Only Memory), and a RAM (Random Access Memory), and functions as an arithmetic device (sequencer) for executing various processes by storing and holding programs for executing various processes described later in a memory (not shown), and executing the programs. In addition, a part or all of each of the processing units 10 to 40 may be configured with other servers (external servers), and the content evaluation system may be realized by distributed processing.

[0049] Next, the processing performed by the content evaluation system will be described in detail.

[0050] 2(A), the evaluation parameter generating unit 30 generates metadata for the content material from the video data of the content material acquired by the content acquiring unit 10. Specifically, in addition to the score when an emotion label of the emotion evoked in the receiver is assigned to each object shown in the video data, narration, subtitles, etc. shown in the video data are extracted as metadata.

[0051] Here, when an emotion label is assigned to each object, scoring is performed for the object's appearance position (from a specific time to a specific time, from specific coordinate xy to specific coordinate xy), for example, scoring the emotion score a for interesting, emotion score b for impactful, etc. (note that emotion scores a, b, etc. are actually between 0 and 10).

[0052] As shown in Figure 2(B), this scoring is performed by calculating an emotion score (scoring value) for each emotion label for each scene and each second of the video. For example, the scoring value of an emotion label such as happy or sad is understood as a transition (change over time) at each point in the video.

[0053] As shown in FIG. 2(A), the content evaluation unit 40 divides the emotion labels generated by the evaluation parameter generation unit 30 into positive elements and negative elements, and evaluates the content material based on the emotion scores of each element included in the positive elements and the emotion scores of each element included in the negative elements (content is not limited to commercials, but may also be programs, videos, movies, slideshows, audio, recordings, television, radio, channels, live filming, live streaming, etc.).

[0054] The emotion labels for positive elements are, as examples, interesting, impactful, exciting, fun, trustworthy, intriguing, warm, enchanting, inspiring, and nostalgic, and each emotion label is associated with an emotion score of a, b, ... j (where a to j are actually a score between 0 and 10).

[0055] In addition to the above, emotional labels for positive elements include exciting, innovative, luxurious, emotional, familiar, soothing, refreshing, stylish, easy to understand, persuasive, relatable, unique, detailed, fresh, diverse, detailed, moving, respectable, admired, cute, likable, favorable, complete, impressive, a sense of accomplishment, and in line with the times.

[0056] The emotion labels for negative elements are, as examples, bored, anxious, persistent, confused, disgust, distrust, fear, sad, embarrassed, and disgusted, and each emotion label is assigned an emotion score k, l,...t (where k to t are actually a score between 0 and 10).

[0057] In addition to the above, other emotional labels for negative elements include banal, rough, suspicious, false, intimidating, unpleasant, disagreeable, discriminatory, derogatory, irritating, too perfect, and out of date.

[0058] Furthermore, not only can positive and negative scores be calculated for combinations within each positive and negative element, but when a score is abnormally high (above a certain threshold) for a combination of multiple items such as positive elements, easy to understand, persuasive, relatable, and luxurious, an emotional score included in the negative elements, such as "too good" or "suspicious," can also be calculated.

[0059] In addition, for each item of the emotion labels of positive and negative elements, there may be an emotion label called "aftertaste of XX." The aftertaste of XX is an index that changes over time with a value that is delayed by a certain time or a variable time with respect to the time progression of XX itself in the content.

[0060] The content evaluation unit 40 analyzes the content material and assigns these emotion labels and scores corresponding to the emotion labels, but not all emotion labels are necessarily assigned. In this case, the content may be characterized only by the emotion labels assigned to the content, without necessarily taking into account the emotion scores.

[0061] Specifically, the content evaluation unit 40 performs all or part of the following evaluation process.

[0062] First, as shown in FIG. 3(A), the content evaluation unit 40 evaluates the content material based on a total score of the emotion scores obtained by adding up the emotion scores of each element included in the positive elements as + and each element included in the negative elements as -.

[0063] In FIG. 3(A), the content evaluation unit 40 calculates the total score for the commercial material data “Aaaa” as content material as 43 points, for “Iiiii” as 61 points, and for “Uuuu” as 37 points.

[0064] The above total score uses the emotion scores at a specific time point (scene) or for the entire commercial material, but instead, emotion scores for every second may be tallied.

[0065] That is, Σ(aT~jT)-Σ(kT~tT) may be calculated and used as the total score, where T is time (seconds).

[0066] In the evaluation process of FIG. 3(A), the final score calculated will be, for example, a score in the quality evaluation of a commercial, and is expected to be used as one of the indicators in advertising transactions. Here, each emotion score is expressed as the sum of the entire content as a total score, Σ(certain emotion label score)T, but the maximum value max(certain emotion label score)T within the content may also be used. In general, a representative value calculated from the time transition of the score of a certain emotion label within the content may be adopted and used to calculate the score of the content.

[0067] In this way, the total score can be calculated using the entire video, an accumulation per second, or a partial item of the emotion score or a score at a partial point in time, and can be calculated as a score of 1 or more.

[0068] Next, as shown in Fig. 3(B), the content evaluation unit 40 evaluates the content material using an impression index based on the emotion label of the element with the highest emotion score from among the emotion scores of each element included in the positive elements and the emotion scores of each element included in the negative elements. This makes it possible to categorize (label, tag) multiple content materials using an impression index such as "soothing," and distribute the content to existing video distribution services.

[0069] In Fig. 3(B), for the content "Aaaa", the emotion labels of fun, excitement, confusion, interest, and persistence are extracted as impression indices, in order from the top, as the top elements of the emotion score. Similarly, for the content "Nice", emotion labels of two elements, trustworthy and bored, are extracted as impression indices. Furthermore, for the content "Uuu", emotion labels of four elements, warmth, nostalgia, anxiety, and sadness, are extracted as impression indices.

[0070] The emotion score used is the emotion score at a specific time point (scene) or the entire commercial material. Alternatively, the impression index may be extracted using a value obtained by aggregating emotion scores for each second.

[0071] In addition, in the present embodiment, the case has been described where emotion labels are extracted in order from the highest emotion score. Alternatively, emotion labels extracted in order from the largest difference compared to the norm (average value) may be used as impression indices.

[0072] In the evaluation process of FIG. 3(B), it is possible to determine what emotion the content particularly evokes, based on the emotion score that is finally calculated.

[0073] Next, the content evaluation unit 40 extracts the emotion labels of the elements with the highest emotion scores that change most per unit time and the time points in question from the time progression shown in FIG. 2(B), i.e., the time progression of the emotion scores of each element included in the positive elements and the emotion scores of each element included in the negative elements, as the starting points of emotion change, and evaluates the content material based on the starting points of emotion change.

[0074] As a result, in FIG. 2(B), the content evaluation unit 40, for example, does not insert an advertisement during the time period when the score for "interesting" is high, since there is no starting point of an emotional change, but extracts the time period when the score for "boring" is high (around 5 minutes in the figure) as the starting point of an emotional change, and inserts an advertisement at this timing, thereby utilizing the score of the emotional transition so as not to impair the viewer's viewing experience or viewing sensation.

[0075] The starting point of the emotion change may be extracted not only from the gradient of the emotion score per unit time, but also from the point when the emotion score is high (the point at which the gradient changes from positive to zero or from positive to negative), or from the maximum value.

[0076] Furthermore, when multiple emotion labels are extracted as emotion change starting points, a combination of multiple emotion labels (emotion scores) may be considered.

[0077] Furthermore, it is preferable to insert commercials with the same emotion label as the inserted content extracted as the impression index in the above embodiment as an ad insertion. By presenting commercials with the same emotion label as the video or program, it is expected that the commercials will be effective in keeping with the context of the video flow (avoiding viewers abandoning the video), and that the viewing effect will be in line with the context.

[0078] Furthermore, emotion labels may be assigned in combination with attributes such as gender and age, and a score may be assigned. For example, to the original emotion label "fresh," new emotion labels such as "fresh (female in her teens)" and "fresh (male in his 40s)" may be assigned. In this way, for example, the time progression of the emotion scores of "fresh (female in her teens)" and "fresh (male in his 40s)" may take different progressions (waveforms) even when analyzing the same content. That is, for example, in the case of video distribution of content on the Internet, different points (timings) may be extracted for the extraction point of the emotion change starting point and the emotion labels of commercials that may be inserted at the emotion change points discussed above, and different commercials (targeting commercials) may be inserted.

[0079] Furthermore, when assigning emotion labels to objects within content, for example, if an anime character from the Showa era is intended for a man in his 40s, the emotion label “nostalgic (man in his 40s)” may be assigned, and if it is intended for a teenage woman, the emotion label “fresh (woman in her teens)” may be assigned, and a high score may be assigned.

[0080] Next, as shown in FIG. 4(A), the content evaluation unit 40 extracts the emotion score portion that exceeds a specific threshold from the time progression of the emotion score of each element included in the positive elements and the emotion score of each element included in the negative elements, together with an emotion label corresponding to the emotion score.

[0081] FIG. 4(A) shows an example of a case where you want to give the thumbnail an impact. The scene with the highest score for “impactful” can be extracted and used for creating thumbnails.

[0082] Regarding the threshold, the user may set a threshold based on the emotion score of the emotion label, and although the maximum and minimum values ​​differ depending on the emotion label, the emotion score may be normalized when setting the threshold. Also, here, a combination of multiple emotion labels (emotion scores) may be considered.

[0083] Note that extracting thumbnails is just one example. For example, when playing a video using video player software, the extracted point may be displayed on the playback seek bar to make it easier to jump the playback (pointer).

[0084] Next, as shown in FIG. 4(B), the content evaluation unit 40 extracts emotion score portions that exceed a specific threshold (in addition to FIG. 4(A)), and links the extracted emotion score portions together.

[0085] FIG. 4(B) shows an example of a case where impact is emphasized when shortening long content. Multiple scenes with high "impactful" emotion scores are extracted and linked together.

[0086] In addition, the cutting position for shortening the video may be set not only to the point where the "impactful" score crosses the threshold as the in point / out point, but also to x seconds before the threshold is exceeded as the in point, y seconds after the threshold is exceeded as the out point, or extracted from a scene change point outside the in-out points, etc., to make the shorter parts easier to see when they are joined together. As explained above in detail, the emotion score may be subjected to a time moving average (smoothing) process in order to remove noise in the changes over time.

[0087] Furthermore, the extraction of the portion exceeding the threshold value may include the portion before and after the portion exceeding the threshold value, and the aforementioned representative value may be used instead of the threshold value.

[0088] As described above in detail, the content evaluation system of this embodiment can comprehensively evaluate the content material itself from multiple perspectives by unraveling the intertwining of a wide variety of positive and negative elements. [Explanation of symbols]

[0089] 10...content acquisition unit, 20...survey data acquisition unit, 30...evaluation parameter generation unit, 40...content evaluation unit, DB1...content database, DB2...survey data database.

Claims

1. A content evaluation system for evaluating content material based on emotions evoked by the content material in a recipient, comprising: an evaluation parameter generation unit that generates an emotion evaluation index by visualizing the emotion evoked in the receiver by the content material using an emotion visualization technique, the emotion evaluation index being an emotion label that is an emotion evoked in the receiver and an emotion score that is a score for the emotion label; a content evaluation unit that evaluates the content material based on the emotion evaluation index generated by the evaluation parameter generation unit; Equipped with The content evaluation unit is When dividing the emotion labels generated by the evaluation parameter generation unit into positive elements and negative elements and evaluating the content material from the emotion scores of each element included in the positive elements and the emotion scores of each element included in the negative elements, A content evaluation system characterized by evaluating the content material using an impression index based on the emotional labels of the elements with the highest emotional scores from among the emotional scores of each element included in the positive elements and the emotional scores of each element included in the negative elements.

2. 2. The content rating system according to claim 1, The content evaluation system, wherein the content evaluation unit classifies the content material using the impression index determined by the content evaluation unit.

3. A content evaluation system for evaluating content material based on emotions evoked in a recipient by the content material, comprising: an evaluation parameter generation unit that generates an emotion evaluation index by visualizing the emotion evoked in the receiver by the content material using an emotion visualization technique, the emotion evaluation index being an emotion label that is an emotion evoked in the receiver and an emotion score that is a score for the emotion label; a content evaluation unit that evaluates the content material based on the emotion evaluation index generated by the evaluation parameter generation unit; Equipped with the content evaluation unit extracts, from the time progression of the emotion score, the emotion labels of the top elements of the emotion score that have the largest changes per hour and the points in time at which the changes were largest as emotion change starting points, and evaluates the content material using the emotion change starting points.

4. A content evaluation system for evaluating content material based on emotions evoked in a recipient by the content material, comprising: an evaluation parameter generation unit that generates an emotion evaluation index by visualizing the emotion evoked in the receiver by the content material using an emotion visualization technique, the emotion evaluation index being an emotion label that is an emotion evoked in the receiver and an emotion score that is a score for the emotion label; a content evaluation unit that evaluates the content material based on the emotion evaluation index generated by the evaluation parameter generation unit; Equipped with the content evaluation unit extracts, from a time progression of the emotional score, a portion of the content material corresponding to the emotional score portion exceeding a specific threshold, together with the emotional label corresponding to the emotional score.

5. 5. The content evaluation system according to claim 4, A content evaluation system characterized in that the content evaluation unit extracts parts of the content material corresponding to the emotional score parts that exceed the specific threshold, and links the extracted parts of the content material together.

6. 6. The content evaluation system according to claim 1, A content evaluation system, characterized in that the evaluation result by the content evaluation unit is provided to an advertisement transaction system.

Images