Information processing apparatus, information processing method, and program

Abstract

To provide an information processor, an information processing method and a program for more effectively distributing a moving image related with betting to be performed by predicting the result of a predetermined game to a user.SOLUTION: An information processing device for distributing, to a user terminal, a video related to a vote cast in anticipation of a result of a predetermined competition, the information processing device including a processor, wherein the processor is configured to identify user identification information associated with the user terminal, select some videos from among a plurality of videos as candidate videos for distribution based on a criterion set in accordance with the user identification information, cause a display unit of the user terminal to display a list showing the selected candidate videos for distribution, and cause the display unit of the user terminal to display one video selected from among the candidate videos for distribution included in the list.SELECTED DRAWING: Figure 12

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and a program. [Background technology]

[0002] Users can post videos over the network and watch videos posted by other users. Today, technologies have been developed that automatically classify a huge number of videos and allow users to select videos to distribute from a predetermined classification, as well as technologies that can extract and distribute videos that are likely to interest viewers (see Patent Documents 1 and 2). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2011 / 126134 [Patent Document 2] Japanese Patent Application Laid-Open No. 2002-320214 Summary of the Invention [Problem to be solved by the invention]

[0004] An example of a distributed video is a content video related to betting (betting) on ​​the outcome of a predetermined competition. When distributing such a video to users, it is preferable to be able to distribute it to users more effectively, and this need increases as the number of videos posted increases.

[0005] Therefore, an object of the present invention is to provide an information processing device, an information processing method, and a program that can more effectively deliver to users videos related to voting that predicts the results of a specified competition. [Means for solving the problem]

[0006] An information processing device according to one aspect of the present invention is an information processing device that distributes videos related to voting to predict the results of a specified competition to users, and is characterized by comprising a selection unit that selects some videos from a plurality of videos as candidate videos for distribution based on specified criteria, and a list transmission unit that transmits a list showing the selected candidate videos for distribution to the user. In addition, an information processing device according to one aspect of the present invention is an information processing device that distributes videos related to voting to predict the results of a specified competition to a user terminal, and is equipped with a processor, wherein the processor identifies user identification information associated with the user terminal, the processor selects some videos from a plurality of videos as candidate videos for distribution based on criteria set according to the user identification information, the processor displays a list showing the selected candidate videos for distribution on a display unit of the user terminal, and the processor displays one video selected from the candidate videos for distribution included in the list on the display unit of the user terminal. [Effects of the Invention]

[0007] According to one aspect of the present invention, when distributing videos related to voting, some videos are selected from multiple videos according to predetermined criteria as candidates for distribution and presented to the user, thereby enabling the videos to be distributed to the user effectively. [Brief explanation of the drawings]

[0008] [Figure 1] FIG. 10 is a diagram showing a display screen of a video related to voting. [Figure 2] FIG. 1 illustrates an information processing apparatus and a user terminal according to an embodiment of the present invention. [Figure 3] FIG. 2 is a diagram illustrating functions of an information processing apparatus according to an embodiment of the present invention. [Figure 4] FIG. 2 is an explanatory diagram of user-related information. [Figure 5] FIG. 2 is an explanatory diagram of video related information. [Figure 6] FIG. 10 is an explanatory diagram of voting-related information. [Figure 7] FIG. 2 is an explanatory diagram of race-related information. [Figure 8] FIG. 10 is an explanatory diagram of hit result information. [Figure 9] FIG. 10 is a diagram showing a display screen of a distribution candidate list. [Figure 10] FIG. 1 is a diagram showing an information processing flow using an information processing apparatus according to an embodiment of the present invention. [Figure 11] FIG. 10 is a diagram showing the flow of a selection process. [Figure 12] FIG. 10 is a diagram showing the flow of a selection process in one embodiment of the present invention. [Figure 13] FIG. 10 is an explanatory diagram of terminal information. [Figure 14] FIG. 10 is a diagram showing another example of user-related information. [Figure 15] FIG. 10 is a diagram illustrating functions of an information processing device according to another embodiment of the present invention. [Figure 16] FIG. 10 is a diagram showing the flow of a selection process in another embodiment of the present invention. [Figure 17] FIG. 10 is a diagram showing the flow of a selection process in another embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0009] An information processing device, an information processing method, and a program according to the present invention will be described in detail below with reference to specific embodiments (first to fourth embodiments). The embodiment described below is merely an example given to facilitate understanding of the present invention, and is not intended to limit the present invention. That is, the present invention may be modified or improved from the embodiment described below without departing from the spirit of the present invention. Furthermore, the present invention includes equivalents thereof.

[0010] The screen examples shown in the drawings referenced in the following description are merely examples, and the screen configuration examples, the content of the information displayed, and the GUI (Graphical User Interface), etc. can be freely designed and changed according to the system design specifications and user preferences, etc.

[0011] In this specification, the term "device" includes not only one device that performs a predetermined function on its own, but also multiple devices that are separate from each other but work together to perform a predetermined function.

[0012] In this specification, the term "user" refers to a user of the information processing device of the present invention, specifically, a user of the service provided by the information processing device of the present invention, and more specifically, a potential recipient of videos distributed by the information processing device of the present invention. Note that, unless otherwise specified, a user is an individual, but may also be a group of multiple people, and is the unit to which a video distribution service account is assigned.

[0013] [Regarding the video distributed in this invention] Before describing the present invention, a description will be given of the video to be distributed in the present invention. In the present invention, the video to be distributed is distributed to users via a communication network such as the Internet or a mobile communication network, and is distributed in real time (live distribution) by streaming, for example. However, the present invention is not limited to this, and can also be applied to cases where recorded (pre-recorded) video is downloaded and played back.

[0014] In the present invention, the video is distributed in a digitized state in a predetermined file format. Note that the specific means used for shooting (recording), digitizing, and distributing the video are not described here because known techniques can be used.

[0015] In the present invention, a "voting support video" is distributed as a video. The voting support video is a video related to voting by predicting the outcome of a predetermined competition, and is distributed to support users in voting. Specifically, the voting support video features at least one performer. During video filming, the performers predict the outcome of the predetermined competition before it is held, actually cast their votes, and check whether their predictions were correct while watching the predetermined competition on TV or the like. Users can watch the voting support video and use it as a reference when casting their votes.

[0016] A "predetermined competition" is, for example, a competition in which you predict the outcome and bet, and if your prediction is correct you will receive a dividend (payout), i.e., a publicly managed competition (race), and specific examples include horse racing, bicycle racing, boat racing, and auto racing. In addition to purchasing a betting ticket by paying a betting amount, voting also includes casting a vote by casting a pseudo-currency (betting value) that can be used in a voting app, which will be described later. The predetermined competition is not limited to publicly managed competitions, but may be any competition for which users can predict the outcome and place bets. The predetermined competition may be, for example, a sports match that is the subject of betting, such as dog racing, rugby, baseball, basketball, soccer, etc., or a gaming tournament such as so-called e-sports, or other competitions in which winners and losers compete against each other.

[0017] In this specification, it is assumed that races are held periodically according to a schedule, for example, every day. The voting support video is distributed with different content each day. To explain in more detail, each day's voting support video is a video that captures the conversations and actions of participants related to the race being held on that day, and can be distributed and viewed only on that day. Note that races are not limited to being held every day, but may also be held every few days, every week, every few weeks, every month, or every few months, or may be held irregularly.

[0018] Furthermore, races are held repeatedly within a predetermined period, for example, multiple times a day, and users can vote for each race during that day. Explaining this in more detail, the time periods during which races (predetermined competitions) are held and the time periods during which voting is accepted alternate throughout the day. During each of these time periods, users can vote for the corresponding race, specifically the race that will be held immediately after that time period. In the voting support video, the cast members vote at least twice a day. The predetermined period is not limited to one day, but may be one to several hours, several days, one to several weeks, one to several months, or one to several years. Also, the number of races held within the predetermined period may be only one, in which case users can vote for one race within the predetermined period.

[0019] Multiple voting support videos can be distributed daily, and each voting support video features a different cast member. Each voting support video features at least one cast member; for example, as shown in FIG. 1, multiple cast members form a group, and a voting support video is filmed and distributed on a group-by-group basis. Note that in a group appearing in a voting support video, each member (each cast member belonging to the group) may jointly predict the outcome of the race and cast the same vote among themselves, or each member may cast a different vote among themselves.

[0020] Furthermore, multiple voting support videos distributed on the same day are all videos about races held on the same day (predetermined competitions held on the same day), and strictly speaking, the content deals with voting for the race. Users can watch at least one of the multiple voting support videos distributed on a given day, and, based on the words and actions of the performers (more specifically, performer groups) in the video they watched, predict the outcome of the race held on that day and cast their vote.

[0021] As shown in Figure 1, the voting support video also displays the details of the votes cast by the cast of the video on that day (in other words, the day the video was released) and information about the votes that were correct. Specifically, the details of the vote, the amount of the vote, the number of hits, the hit rate, the amount of the hit, and the number of wins are displayed. Users can refer to the above displayed information while watching the voting support video, and can, for example, cast a vote with the same prediction as the cast of the video they are watching, i.e., piggyback vote.

[0022] The "betting content" refers to the predicted content of the bet, specifically the predicted content of the finishing order in the race. More specifically, the betting content corresponds to the combination of betting types such as quinella, trifecta, trifecta, etc. and the predicted finishing order corresponding to that type. The runners in the race whose finishing order is predicted may also be a combination of people and vehicles (for example, horses, bicycles, motorcycles, or boats). "Vote amount" is the amount of money cast in a vote (including the amount cast using pseudo-currency). The "number of hits" refers to the number of votes for which the predictions were correct in one race, for example, the number of hit votes in the immediately preceding race. Note that if multiple races are held in one day, the "number of hits" may be the total number of hit votes up to that point in time on that day. The "hit rate" is the percentage of votes for which predictions were correct during a specified period. For example, if multiple races are held in one day, it is the hit rate at the current point in time for that day. Note that the hit rate may also be calculated from the total number of votes cast by the cast of the voting support video during the distribution of the video, including previously distributed videos, and the number of correct votes among those total. The "winning amount" is the payout (refund) for a bet that predicts a race correctly, for example, the payout for the most recent race. If multiple races are held in one day, the total amount of winning amounts up to that point on that day may be used as the "winning amount." The "number of wins" refers to the number of races in which predictions were correct. For example, if multiple races are held in a single day, it refers to the total number of correct predictions for that day. Note that the "number of wins" may also refer to the total (cumulative) number of races in which the performer of the voting support video has made correct predictions in all of the videos he or she has distributed so far, including videos distributed in the past.

[0023] The voting support video can be viewed using dedicated application software (hereinafter referred to as the voting app). That is, the voting app is installed on the user's device (hereinafter referred to as the user terminal), and when the user performs a predetermined operation after launching the app, the user terminal receives and unpacks the distribution data of the voting support video. As a result, the voting support video is displayed on the display (display unit) of the user terminal.

[0024] [Regarding the information processing device and user terminal of the first embodiment] An information processing device according to a first embodiment of the present invention (hereinafter referred to as information processing device 10) and a user terminal used by a user (hereinafter referred to as user terminal 12) will be described.

[0025] In the following description, a "viewer" refers to a user who is watching or has watched a voting support video. For ease of understanding, the following description will be based on the assumption that a voting support video is distributed to one user (hereinafter, user γ). However, the present invention is also applicable to cases where a voting support video is distributed to multiple users.

[0026] The information processing device 10 is composed of a server computer (corresponding to an example of a computer) and distributes voting support videos to user γ. More specifically, it communicates with a user terminal 12 used by user γ and executes information processing for distributing the voting support videos. The server computer constituting the information processing device 10 may be a single computer or multiple computers distributed in parallel. Furthermore, the server computer may be a server computer for an ASP (Application Service Provider), SaaS (Software as a Service), PaaS (Platform as a Service), or IaaS (Infrastructure as a Service).

[0027] 2, the information processing device 10 and the user terminal 12 of user γ are communicatively connected via a communication network 14, and cooperate with each other to form an information processing system S. The communication network 14 may include a LAN (Local Area Network), a WAN (Wide Area Network), an intranet, Ethernet (registered trademark), and the like.

[0028] In the information processing system S, a server computer constituting an information processing device 10 performs a series of information processing related to the present invention (excluding input and display of information), including video distribution. On the user terminal 12 side, information to be handed over to the server computer is input, and information distributed from the server computer is output (display, playback, etc.)

[0029] The information processing device 10 acquires voting support videos posted from multiple groups, analyzes each video, and distributes voting support videos that meet predetermined conditions based on the analysis results to user γ. The information processing device 10 also accepts votes for races submitted by users via a voting app, acquires the race results, and determines whether the predictions for each accepted vote were correct. Furthermore, the information processing device 10 calculates the number of hits, number of wins, winning amount, and hit rate (hereinafter referred to as hits, etc.) for each accepted vote based on the results of the prediction hit determination. Furthermore, the information processing device 10 distributes information regarding the calculated hits, etc. for the performers in the voting support video to user γ along with the voting support video.

[0030] User γ's user terminal 12 is composed of a personal computer, smartphone, mobile phone, tablet terminal, or other terminal with communication functions, and is used by user γ to view voting support videos distributed from the information processing device 10, predict the results of the race, and place votes.

[0031] Specifically, the user terminal 12 receives data sent from the information processing device 10 using the installed voting app, expands the data, and displays various information on the display (corresponding to a display unit) of the user terminal 12. The data sent from the information processing device 10 includes information about the voting support video and the performers in the video. The information about the performers in the video includes profiles of the performers (specifically, each member of the group appearing in the video), voting information about votes cast by the performers during the video distribution (specifically, the race for which the vote was cast, the content of the vote, the amount of the vote, etc.), and the number of winning combinations of the performers, etc.

[0032] Referring to FIG. 1, the display screen of the user terminal 12 while the app is running will be described. A voting support video is displayed in the upper region T1 of the display. In addition, together with the display of the voting support video, information about the performers in the voting support video is displayed in a central region T2 different from the upper region T1. The information displayed in the central region T2 is information associated with the voting support video displayed in the upper region T1, and includes voting information. In this way, the voting support video and voting information are displayed on the display at one time (within one screen).

[0033] Additionally, a voting button Bt1 is displayed in the lower region T3 located at the bottom of the display. User γ can cast a vote for a race by clicking the voting button Bt1 and entering the details of the vote on a voting screen (not shown). Here, the upper region T1 corresponds to the first display region of the display, and the central region T2 and the lower region T3 correspond to the second display region of the display. Note that instead of the voting button Bt1, a field for entering the details of the vote may be displayed in the lower region T3.

[0034] During distribution of a voting support video, if a cast member in the video predicts the outcome of the next race (for example, the race to be held immediately after) and casts a vote, the prediction, the betting amount, and other voting information are displayed in the central area T2. At this time, a piggyback vote button Bt2 is displayed in the lower area T3. In this case, user γ can click the piggyback vote button Bt2 to cast a vote with the same prediction as the vote cast by the applicant on the voting support screen they are viewing, i.e., to cast a piggyback vote.

[0035] After the race is over, information about the race results, etc. is displayed on the user terminal 12, although the screen will be different from that shown in Fig. 1. This allows user γ to check whether the predictions related to his / her bets were correct or not, and the winning amount (payout) if the predictions were correct.

[0036] [Configuration of information processing device] Next, the configuration of the information processing device 10 will be described. As shown in FIG. 2, the server computer that constitutes the information processing device 10 has, as hardware devices, a processor 21, a memory 22, a communication interface 23, and a storage 24, and these devices are electrically connected via a bus 25.

[0037] Furthermore, the server computer has installed therein, as software, a program for an operating system (OS) and an application program for video distribution. These programs correspond to the "programs" of the present invention. When the processor 21 operates in accordance with the programs, the server computer functions as the information processing device of the present invention and executes a series of processes related to the present invention. The above program may be obtained by reading it from a computer-readable recording medium, or may be obtained by receiving (downloading) it via a network such as the Internet or an intranet.

[0038] The processor 21 may be configured by a CPU (Central Processing Unit), MPU (Micro-Processing Unit), MCU (Micro Controller Unit), GPU (Graphics Processing Unit), DSP (Digital Signal Processor), TPU (Tensor Processing Unit), or ASIC (Application Specific Integrated Circuit), etc. The memory 22 may be configured by semiconductor memories such as a ROM (Read Only Memory) and a RAM (Random Access Memory).

[0039] The communication interface 23 may be configured by, for example, a network interface card, a communication interface board, etc. The standard of data communication by the communication interface 23 is not particularly limited, and examples include communication by a wireless LAN based on Wi-fi (registered trademark), communication by a 3G to 5G or later generation mobile communication system, or communication based on LTE (Long Term Evolution).

[0040] The storage 24 may be configured with a flash memory, a hard disc drive (HDD), a solid state drive (SSD), a flexible disc (FD), a magneto-optical disc (MO disc), a compact disc (CD), a digital versatile disc (DVD), a secure digital card (SD card), a universal serial bus memory (USB memory), or the like. The storage 24 may be built into the housing of a server computer constituting an information processing device, or may be externally attached to the server computer. The storage 24 may also be configured with an external computer (e.g., a database server) communicably connected to the server computer. A distributed ledger technology such as a blockchain may be used as a technology for recording various data in order to prevent unauthorized data tampering, etc.

[0041] The configuration of the information processing device 10 will be explained again from a functional perspective. The information processing device 10 has functional units shown in Fig. 3, namely, a video acquisition unit 31, a vote acquisition unit 32, a result acquisition unit 33, a memory unit 34, an update unit 35, a selection unit 36, a list transmission unit 37, and a video distribution unit 38. Of these functional units, the memory unit 34 is realized by the memory 22 or the storage 24, and the other functional units are realized by cooperation between various hardware devices, including the processor 21, of the server computer that constitutes the information processing device 10 and programs installed on the server computer. Each functional unit will be described in detail below.

[0042] The video acquisition unit 31 acquires the posted voting support videos on a group basis (i.e., for each performer group). In the first embodiment, since the voting support videos are distributed live, the video acquisition unit 31 acquires the voting support videos being shot in real time (strictly speaking, including delay times that occur during data communication).

[0043] Voting support videos are posted with the video's ID number, distribution date and time, etc., incorporated, and are also linked to information about the performers of the video (more specifically, the performer group and its members). The information about the performers includes the performer's attributes (category), specifically, the gender, age, occupation, etc. of each member of the performer group. The performer's attributes (category) may be determined by the performer themselves, or may be determined by the information processing device 10 in accordance with predetermined rules.

[0044] When a user casts a vote for a race, the vote acquisition unit 32 acquires information about the vote from the user terminal 12 that cast the vote. In other words, the vote acquisition unit 32 accepts votes by receiving from the user terminal 12 the voting details (specifically, the prediction details, the betting amount, etc.) that the user inputs through the user terminal 12. Users who cast votes include user γ, as well as performers in the voting support video and viewers of voting support videos that are currently being distributed or have been distributed in the past. In the first embodiment, multiple races are held in one day, and voting is carried out during reception time periods set between races. During each reception time period, the vote acquisition unit 32 acquires information about votes for the race that will be held immediately afterwards.

[0045] Furthermore, the display of the user terminal 12 of user γ displays both the voting support video and the voting information associated with the voting support video (on one screen). This allows user γ to cast a vote based on the voting information while watching the voting support video. The voting information associated with the voting support video is information about the votes cast by the performers in the video, specifically, the predicted content of the vote, the amount of the vote, etc. Then, the vote acquisition unit 32 accepts the vote of user γ based on the voting information through the user terminal 12 of user γ. That is, the vote acquisition unit 32 can accept the vote of user γ by receiving, from the user terminal 12, the voting content that user γ inputs through the user terminal 12 with both the voting support video and the voting information displayed on the display.

[0046] When each race is completed, the result acquisition unit 33 acquires information about the results of the race. The information about the results of the race includes the race ID, the finishing order of each runner who participated in the race, information about the payouts according to the race results (specifically, odds), and other information about the race results.

[0047] The storage unit 34 stores various types of information necessary for information processing by the information processing device 10, including distribution of voting support videos. The information stored in the storage unit 34 includes user-related information, video-related information, voting-related information, race-related information, and winning result information, as shown in FIG.

[0048] User-related information is information about users, including the cast and viewers of the voting support video, and is stored for each user. As shown in Figure 4, the user-related information includes the user's name, user ID corresponding to user identification information, profile, viewing history information, voting history information, winning record information, the amount of in-game currency owned, and the group to which the user belongs. The profile is information about the attributes (categories) of the user, such as gender, age, and occupation. Viewing history information is information about the viewing status of videos, and specifically, information indicating the viewing date and time of voting support videos that the user has viewed, the video ID, and whether or not the user cast a piggyback vote while watching the video. The voting history information is information indicating the date and time of voting, the vote ID, and the like for the votes that the user has voted so far. The hit record information is information that indicates the hit record of the predictions related to the votes made by the user, and specifically includes, for example, the amount of winnings won in the most recent race, as well as the hit rate, number of hits, amount of winnings, and number of wins on the day of the race. The group to which the user belongs is information indicating the performer group to which the user belongs if the user is a performer in the voting support video.

[0049] The video-related information is information about the posted voting support video and is stored for each video. As shown in Fig. 5, the video-related information includes the video ID, distribution date and time, the name of the group appearing in the video, performer information, attributes (category), target race information, and evaluation value. The performer information is information about the group and its members appearing in the video, and specifically includes each member's ID (user ID), the amount of votes each member cast for the race held on the day the video is distributed, as well as the number of wins, number of hits, amount of hits, and hit rate on the day the video is distributed. The attribute (category) is information about the attribute (category) of the group appearing in the video, and is determined according to the attributes (for example, age, sex, occupation, etc.) of the members of the group. The target race information is information about the race (hereinafter referred to as the target race) whose results are predicted in the voting support video, such as the ID of the target race, the date and venue of the target race, etc. The evaluation value is an evaluation value (score) given to the voting support video through video analysis by the information processing device 10. The evaluation value will be described in detail later.

[0050] The voting-related information is information about a vote made by a user and is stored for each vote. As shown in Fig. 6, the voting-related information includes the ID of the vote, the ID of the user who made the vote, the ID of the race related to the vote, the date, time, and location of the race, the type of vote, the predicted content (i.e., the predicted finishing order), the amount of the vote, the payout ratio (odds) if the prediction is correct, and the time of the vote. In addition, if a user makes a piggyback vote, the voting-related information also includes information for identifying the voting support video (hereinafter referred to as the reference video) that the user was watching when the piggyback vote was made, specifically the ID of the reference video.

[0051] Race-related information is information about a race, and is stored for each race. As shown in FIG. 7, the race-related information includes the race ID, event, race grade (rating), date and time of the race, venue, runner information, and voting candidate information. The race event corresponds to a competition event such as auto racing or bicycle racing. The voting candidate information is information about predicted finishing order candidates in a race, and as shown in FIG. 7, includes the candidate ID, voting details (specifically, the details of the predicted finishing order), odds, and the like.

[0052] The winning result information is information indicating the winning result of the race result predicted by the user at the time of voting, and is stored for each vote. As shown in Fig. 8, the winning result information includes the vote ID, the ID of the user who cast the vote, the ID of the race that the vote was about, whether the prediction was correct or not, the amount of the dividend won if the prediction was correct (winning amount), and the payment deadline for the dividend (refund deadline).

[0053] The update unit 35 appropriately updates the information stored in the storage unit 34. For example, when the video acquisition unit 31 acquires a new voting support video, the update unit 35 adds video-related information about the newly acquired voting support video to the storage unit 34. Furthermore, when the result acquisition unit 33 acquires information about the race results of a certain race, the update unit 35 reads out from the storage unit 34 voting-related information about the votes cast for the race, and determines whether or not each vote was a hit based on the information acquired by the result acquisition unit 33. Furthermore, the update unit 35 updates the hit result information and user-related information based on the determination result, and updates, for example, the number of wins, the number of hits, and the winning amount of the users whose predictions were correct.

[0054] When multiple voting support videos are posted and acquired by the video acquisition unit 31, the selection unit 36 ​​selects some voting videos from the multiple voting support videos as candidate videos to be distributed based on predetermined criteria. Candidate videos to be distributed are candidates for voting support videos to be distributed to user γ, and in the first embodiment, for example, three videos are selected as candidate videos to be distributed. Note that the number of voting support videos selected as candidate videos to be distributed is not particularly limited, as long as at least one video is selected.

[0055] The predetermined criteria are criteria (conditions) for selecting candidate videos for distribution. For example, the criteria are set for the evaluation results when evaluating each of a plurality of voting support videos. In other words, the selection unit 36 ​​selects candidate videos for distribution based on the evaluation results of each video. In particular, in the first embodiment, the selection unit 36 ​​calculates an evaluation value for each voting support video and selects candidate videos for distribution based on the magnitude of the evaluation value.

[0056] Specifically, as shown in FIG. 3, the selection unit 36 ​​includes an analysis unit 36A and an evaluation value calculation unit 36B. The analysis unit 36A analyzes each of the multiple voting support videos acquired by the video acquisition unit 31 (strictly speaking, real-time voting support videos currently being shot) based on the audio and images (video) in the video and various information stored in the storage unit 34. In the first embodiment, the analysis unit 36A analyzes each voting support video from the perspective of the degree of engagement of the performers in the video, the status of voting by the performers for races held on the video distribution date, and the size of the viewers who support the performers. The degree of engagement is determined, for example, based on the performers' words, actions, or facial expressions. The status of voting is determined, for example, by the amount of bets and winning results. The size of the viewership corresponds to, for example, the number of viewers who cast piggyback votes.

[0057] In the first embodiment, multiple races are held in one day (predetermined period). When votes are cast for each race, the update unit 35 updates the information stored in the memory unit 34 as needed throughout the day. In addition, the analysis unit 36A analyzes the voting support video every time a race is held in one day and updates the analysis results.

[0058] The evaluation value calculation unit 36B evaluates each of the multiple voting support videos, and specifically calculates an evaluation value for each voting support video based on the analysis results by the analysis unit 36A. In the first embodiment, the evaluation value calculation unit 36B calculates the evaluation value for each voting support video from the three perspectives described above. In the first embodiment, as described above, the analysis results for each voting support video are updated as needed within one day (predetermined period). When the analysis results are updated, the evaluation value calculation unit 36B recalculates the evaluation value for each voting support video based on the updated analysis results.

[0059] Then, the selection unit 36 ​​selects candidate videos for distribution based on the evaluation results (i.e., evaluation values) of each voting support video, and specifically, selects the top three voting support videos with the highest evaluation values ​​as candidate videos for distribution.

[0060] The list transmission unit 37 transmits to user γ a ​​list indicating the distribution candidate videos selected by the selection unit 36 ​​(i.e., voting support videos with the first to third highest evaluation values). Specifically, the list transmission unit 37 reads information (e.g., video-related information, user-related information, etc.) stored in the storage unit 34 regarding the voting support videos selected as distribution candidate videos. The list transmission unit 37 generates list display data based on the read information. The generated list display data is transmitted to the user terminal 12 of user γ, and is displayed in the user terminal 12 of user γ after receiving the list display data. As a result, the distribution candidate list shown in FIG. 9 is displayed on the display of the user terminal 12 of user γ. That is, the list transmission unit 37 displays the distribution candidate list on the display (display unit) of the user terminal 12.

[0061] The distribution candidate list displays selectable image information for each of the three selected distribution candidate videos. The image information for the video may be a thumbnail image created based on a portion of the distribution candidate video (specifically, a frame image), or may be a photograph of a cast member of the distribution candidate video. User γ selects one of the distribution candidate videos and touches the image information of the selected video in the displayed distribution candidate list.

[0062] In the distribution candidate list, the placement positions of the three selected distribution candidate videos may be set according to the evaluation values. More specifically, priorities are set for the three selected distribution candidate videos according to the evaluation values. Then, when displaying the distribution candidate list on the display of the user terminal 12, the list sending unit 37 (in other words, the processor) arranges the distribution candidate videos in an order according to their respective priorities. Specifically, for example, the three distribution candidate videos are arranged in the distribution candidate list so that videos with higher priorities are positioned higher in the list. Note that how the priorities are set, i.e., the setting criteria for setting the priorities, are not particularly limited.

[0063] The video distribution unit 38 distributes (more specifically, distributes in real time) the voting support video selected by user γ from among the distribution candidate videos displayed in the distribution candidate list to the user terminal 12 of user γ. The distributed voting support video is displayed on the display of the user terminal 12. This allows user γ to view the video with the highest evaluation value and selected by user γ himself from among the multiple voting support videos posted.

[0064] As described above, the video distribution unit 38 displays one voting support video selected from the distribution candidate videos included in the distribution candidate list on the display (display unit) of the user terminal 12. At this time, the video distribution unit 38 displays the voting support video in the upper region T1 of the display, and also displays voting information associated with the video in the central region T2 of the display (see FIG. 1).

[0065] [Information processing method according to the first embodiment] Next, as an example of the information processing method of the present invention, an information processing flow (hereinafter, video distribution flow) for distributing a voting support video to user γ using the information processing device 10 will be described. The video distribution flow employs the information processing method of the present invention. In other words, each step in the video distribution flow corresponds to a component of the information processing method of the present invention. The flow described below is merely an example, and unnecessary steps may be deleted, new steps may be added, or the order in which steps are performed may be changed, without departing from the spirit of the present invention.

[0066] In the following, we will assume that multiple races are held on a certain day (hereinafter referred to as a race day), and that voting support videos are distributed in real time on the race day to user γ. In this case, the performers (more specifically, performer groups) in the voting support videos predict the race results for each of the multiple races held on the race day and cast their votes.

[0067] The video distribution flow proceeds according to the flow shown in Fig. 10, and each step in the flow is mainly performed by a processor (hereinafter simply referred to as a processor) of a server computer that constitutes information processing device 10. Here, the execution of each step by a processor is synonymous with the execution of each step by the server computer (computer) that constitutes information processing device 10.

[0068] To explain the video distribution flow, at a predetermined time on the day of the race, multiple performer groups (hereinafter referred to as groups A to Z) each start shooting and posting a voting support video. The voting support video posted by each group corresponds to multiple videos. The processor acquires the voting support video for each group (S001).

[0069] Furthermore, each of the multiple users, including the performers in the voting support video, predicts the outcome of the next race, i.e., the race to be held immediately after, during the voting acceptance period on the race day (i.e., the period between races), and casts a vote according to the prediction. The processor acquires information regarding the votes cast by each user (S002). At this time, the processor determines whether each vote is a piggyback vote, and for piggyback votes, identifies the referring video (S003, S004).

[0070] The processor acquires information on the race results after the end of each race (S005), determines whether or not the predictions for each of the votes made during the immediately preceding reception period were correct, and determines the number of correct votes for the votes whose predictions were correct (S006, S007). At this time, the processor determines, for example, the number of correct votes for the performers in the voting support video, specifically for each of groups A to Z.

[0071] On the other hand, after starting step S001 (i.e., when the processor starts acquiring voting support videos), it executes a selection process (S008). In the selection process, the processor selects a predetermined number (specifically, three) of videos from the voting support videos acquired for each group as candidate videos for distribution.

[0072] After the selection process is performed, the processor transmits a list showing the selected distribution candidate videos, i.e., a distribution candidate list, to user γ (S009). Specifically, the processor reads information about the voting support videos selected as distribution candidate videos from the storage unit 34. Then, the processor generates list display data based on the read information and transmits the generated list display data to the user terminal 12 of user γ. In the user terminal 12 of user γ, the list display data is expanded and the distribution candidate list is displayed (see FIG. 9).

[0073] Thereafter, user γ selects one of the three voting support videos on the distribution candidate list (i.e., a video from one of groups A to Z) and touches it on the list display screen. The processor accepts the video selection by user γ (S010). This triggers the processor to distribute the voting support video selected by user γ to user terminal 12 of user γ, specifically, in real time (S011). In other words, the processor displays one voting support video selected from the distribution candidate videos included in the distribution candidate list on the display (display unit) of user terminal 12.

[0074] The above series of steps S001 to S011 are repeated until a predetermined termination condition is met. The termination condition may be that the final race on the race day has finished, or that user γ has stopped the voting app on user terminal 12 and finished watching the voting support video. Once the termination condition is met, the video distribution flow for the race day ends.

[0075] Next, the selection step in the video distribution flow will be described in detail with reference to FIG. In the selection step, the processor executes an analysis process for each of the voting support videos acquired for each group in step S001, specifically, executes a first analysis process, a second analysis process, and a third analysis process (S021).

[0076] In the first analysis process, the processor determines the degree of activity in each voting support video based on at least one of the audio and image of the performers in the video (i.e., each member of groups A to Z), and more specifically, the behavior of the performers. More specifically, in the first analysis process, the processor determines the following (r1) to (r3) for each of the multiple voting support videos: (r1) Number of people speaking in the video (r2) Content of the performers' speech (topics) (r3) Facial expressions of the performers in the video

[0077] The number of people speaking in a video can be determined based on the voices of the performers in the video using known speaker recognition technology such as voiceprint personal identification technology, etc. The number of people speaking may also be determined using image processing technology that analyzes mouth movements in images of the performers, or may be determined based on both the voices and images of the performers. The content of the performers' speech can be identified by converting the performers' speech into text (character information) using speech recognition processing and natural language processing based on the audio in the video. To identify the facial expressions of performers, performers are extracted from a video using known object analysis algorithms such as Region-based CNN (R-CNN), Fast R-CNN, You Only Look Once (YOLO), and Single Shot Multibox Detector (SDD). Then, based on the images of the extracted performers, facial expressions of the performers can be identified by applying image analysis techniques for facial expression recognition based on FACS theory (Facial Action Coding System).

[0078] In the first analysis process, it is more preferable to identify the degree of activity (specifically, the above items r1 to r3) during the betting acceptance period between races on the day the race is held.

[0079] In the second analysis process, the processor identifies information about the votes cast on the race day by the cast members (i.e., each member of groups A to Z) for each voting support video, specifically identifying the bet amounts and winning results, etc. More specifically, in the second analysis process, the processor identifies the following (r11) to (r13) for each of the multiple voting support videos on a group-by-group basis. (r11) Number of winnings in the race immediately preceding the race date (r12) The number of consecutive votes in which the prediction was correct among multiple votes cast by the cast of the voting support video during the race day. (r13) Change in voting amount for each vote on the race day

[0080] The number of hits, etc., for the immediately preceding race on the race day is the numerical value related to the votes for which the predictions of the video performers were correct in the immediately preceding race. The number of hits, etc., is determined based on information stored in the memory unit 34 after the immediately preceding race ended, specifically, the user-related information, voting-related information, and hit result information of the performers in the voting support video. The number of consecutive votes with correct predictions within a race day is identified based on the video-related information of the voting support video stored in the storage unit 34, the voting-related information, and the correct result information. The change in betting amount on a race day is the increase or decrease in betting amount for each race on the race day, and is determined based on the user-related information of the performers in the voting support video stored in memory unit 34 and the voting-related information.

[0081] In the second analysis process, instead of identifying the number of winning combinations in the immediately preceding race on the race day, or in addition to identifying the number of winning combinations in the immediately preceding race, the cumulative number of winning combinations up to the current point on the race day may be identified.

[0082] In the third analysis process, the processor identifies the size of the audience that supports the performer for each voting support video, specifically, the number of viewers who cast piggyback votes while watching the video for each group. The number of viewers who cast piggyback votes is identified based on the video-related information and voting-related information of the voting support video stored in the memory unit 34. In addition, in the third analysis process, instead of identifying the number of viewers who cast their votes while watching the video, or in addition to identifying the number of viewers who cast their votes, the number of viewers who cast their votes while watching the video may be identified, regardless of whether the content of the votes predicted by the video's performers is different.

[0083] After the analysis process is completed, the processor executes an evaluation process (S022) to evaluate each of the voting support videos acquired for each group in step S001. In the evaluation process, an evaluation value for each voting support video is calculated for each group based on the analysis results of the analysis process.

[0084] The procedure for calculating the evaluation value based on the analysis results will be described below. When the number of people speaking in the video is identified in the first analysis process, if there is more than one person, the evaluation value is set to a higher value. Here, if the number of people speaking is the number of members of the group (i.e., three people), the evaluation value is set to the highest. However, this is not limited to this, and for example, if the number of members of the group is more than three (e.g., ten or more people), the evaluation value may be lowered if the number of people speaking exceeds a predetermined number.

[0085] Furthermore, when the speech content of the performer is identified in the first analysis process, if the speech content contains predetermined words, for example, positive words, the evaluation value should be set to a higher value. Conversely, if the speech content contains negative words, the evaluation value should be set to a lower value. Note that positive and negative words should be stored in advance in the computer as keywords for judgment.

[0086] Furthermore, when the facial expression of a performer is identified in the first analysis process, if the facial expression is a predetermined facial expression, for example, a positive facial expression such as a smile, the evaluation value is set to a higher value. Conversely, if the facial expression is a negative facial expression such as an angry face or a crying face, the evaluation value is set to a lower value. Note that it is advisable to store positive and negative facial expressions in advance in the computer as facial expression patterns for judgment.

[0087] Furthermore, when the number of winning predictions in the immediately preceding race is identified in the second analysis process, the greater the value of the winning predictions, the higher the evaluation value will be. Furthermore, when the number of consecutive bets with correct predictions is identified in the second analysis process, the greater the consecutive number of bets, the higher the evaluation value will be. Furthermore, when the change in bet amount, i.e., the increase or decrease in bet amount, is identified in the second analysis process, the greater the increase in bet amount since the previous race, the higher the evaluation value will be.

[0088] Furthermore, when the number of viewers who have cast a piggyback vote while watching the video is identified in the third analysis process, the greater the number of viewers, the higher the evaluation value will be.

[0089] When calculating the evaluation value, factors other than the analysis results of the analysis process may be taken into consideration. For example, the evaluation value may be calculated based on the current number of viewers of the voting support video, or the cumulative number of viewers on the day of the race.

[0090] In the evaluation process, the processor calculates an evaluation value for each video (in other words, for each performer group) in the above manner. At this time, an evaluation value based on the analysis results of the first analysis process, an evaluation value based on the analysis results of the second analysis process, and an evaluation value based on the analysis results of the third analysis process may be calculated separately, and these three evaluation values ​​may be added together to obtain a final evaluation value. In this case, a weight may be assigned to each evaluation value, and each of the three evaluation values ​​may be multiplied by the corresponding weight, and the products may be added together to obtain a final evaluation value.

[0091] After the evaluation process is completed, the processor selects a predetermined number (e.g., three) of candidate videos to be distributed from among the multiple voting support videos according to the evaluation value calculated for each video (S023). More specifically, the processor selects the predetermined number of voting support videos as candidate videos to be distributed in descending order of evaluation value. However, without being limited to this, for example, multiple voting support videos may be divided by attribute (category), and the voting support video with the highest evaluation value for each attribute may be selected as a distribution candidate video. In this case, the distribution candidate list subsequently sent to user γ may display information (image information) about the distribution candidate videos by category.

[0092] The selection process is repeatedly performed according to the above procedure during the race day; more specifically, it is performed each time a race is held on the race day. In other words, each time a race is held during the race day, an evaluation value is calculated (updated) for each of the multiple voting support videos. Then, distribution candidate videos are re-selected in accordance with the update of the evaluation value for each video. Regarding the distribution candidate list sent to user γ, a new list may be sent each time a distribution candidate video is selected. Alternatively, the distribution candidate list may be sent only once after user γ operates the user terminal 12 to launch the voting app. In this case, the distribution candidate list indicates the distribution candidate videos selected immediately before the app was launched.

[0093] [Regarding the second embodiment] In the first embodiment, a predetermined number of voting support videos are selected as distribution candidate videos in descending order of evaluation value. However, this is not limited to this, and distribution candidate videos may be selected not only based on evaluation value but also based on user γ, who is the recipient of the distribution candidate list. In other words, the selection process may be performed so that distribution candidate videos differ for each user. This embodiment (hereinafter, referred to as the second embodiment) will be described with reference to FIG. 12. Note that the following mainly describes the differences between the second embodiment and the first embodiment.

[0094] The video distribution flow according to the second embodiment is generally the same as the video distribution flow according to the first embodiment. On the other hand, the selection process according to the second embodiment (hereinafter referred to as the second selection process) is triggered when user γ launches a voting app on the user terminal 12. In the second selection process, first, the user ID associated with the user terminal 12 in which the voting app is activated is identified (S101). Specifically, the processor identifies the terminal ID of the user terminal 12 based on data sent from the user terminal 12. The processor then reads the terminal information stored in the storage unit 34 and identifies the user ID associated with the user terminal 12 from the terminal information. In other words, the processor identifies the user ID of user γ who activated the voting app.

[0095] The terminal information is information that associates the terminal ID (terminal identification information) of the user terminal 12 with the name and user ID of the user who owns the user terminal 12, as shown in FIG.

[0096] The processor sets criteria for selecting videos according to the user ID of the identified user (S102). In step S102, the processor sets criteria for selecting videos for each user based on information associated with the user ID, etc. Then, the processor selects some videos from the multiple voting support videos as distribution candidate videos based on the set criteria (S103). In step S103, the processor selects distribution candidate videos for each user based on the criteria for each user.

[0097] As a method for setting criteria for selecting moving images in the second selection process, for example, the following four methods can be mentioned. First method: A method that uses the narrowing conditions specified by user γ as the criteria. Method 2: Setting criteria based on users' voting history information Method 3: Setting criteria based on user viewing history information Fourth method: Setting criteria based on other users' viewing history or voting history

[0098] Regarding the first method, the user whose user ID was identified in step S101, i.e., user γ, can specify filtering conditions through the user terminal 12 associated with the user ID. Filtering conditions are matters that should be prioritized when selecting videos to be distributed, and examples include the popularity of the performers in each video, the number of hits that the performers made on the day of the race, and the number of viewers who voted on the basis of their participation while watching each video. Furthermore, the filtering conditions may include video attributes (categories) or race attributes. Race attributes include the grade, type, event, and venue of the race.

[0099] If the criteria are set using the first method, in step S103, distribution candidate videos are selected based on the evaluation value calculated for each voting support video according to the procedure described above and the filtering conditions specified by user γ. Specifically, when user γ operates user terminal 12 to specify the filtering conditions, the processor communicates with user terminal 12 and receives data indicating the filtering conditions from user terminal 12.

[0100] The processor calculates, for each voting support video, an evaluation value based on the analysis results of the first analysis process, an evaluation value based on the analysis results of the second analysis process, and an evaluation value based on the analysis results of the third analysis process.The processor then selects candidate videos for distribution based on the evaluation values ​​calculated for each analysis process that correspond to the filtering conditions.For example, if user γ specifies the popularity of the performers in the video as a filtering condition, the selection process selects candidate videos for distribution based on the evaluation value based on the analysis results of the first analysis process (i.e., the popularity).

[0101] As described above, when the criteria are set using the first method, candidate videos for distribution can be selected taking into account the narrowing conditions specified by user γ. This allows candidate videos for distribution to be selected in a way that reflects user γ's preferences or requests.

[0102] In the second method, the processor reads out information about the user γ whose user ID was identified in step S101, that is, the voting history information among the user-related information, from the storage unit 34. As can be seen from FIG. 4, the voting history information is information associated with the user ID.

[0103] The processor identifies a predetermined value from the read voting history information. The predetermined value is, for example, a value calculated from information associated with the voting ID indicated in the voting history information, specifically the number of votes or the hit rate of the prediction by user γ. This value is calculated for each attribute of the race based on user-related information including the voting ID, voting-related information, and race-related information. The processor also identifies the attributes of the race for which the above values ​​satisfy a predetermined condition, specifically the attributes (grade, type, event, race venue, etc.) of the race for which user γ's number of votes or hit rate reaches a threshold. This identified race attribute is also referred to below as the "target attribute."

[0104] The processor extracts multiple voting support videos related to voting in races of the target attribute from multiple voting support videos that can be distributed at that time. The processor also performs the first to third analysis processes on the extracted multiple voting support videos using the procedure described above, and calculates an evaluation value for each video based on the analysis results of each analysis process. The processor then selects a distribution candidate video from the multiple voting support videos related to voting in races of the target attribute according to the evaluation value calculated for each video.

[0105] As described above, when the criteria are set using the second method, candidate videos for distribution can be selected taking into account the voting history of user γ. This allows candidate videos for distribution to be selected that reflect the attributes of races that user γ prefers or the attributes of races for which user γ's predictions have a high hit rate.

[0106] When selecting candidate videos for distribution based on evaluation values ​​based on the analysis results of the first to third analysis processes, voting support videos for races of the target attribute may be selected preferentially. In this case, several voting support videos for races of attributes other than the target attribute may be added to the candidate videos for distribution.

[0107] There are other possible cases where criteria are set using the second method. For example, the processor may identify the number of piggyback votes that user γ has cast in the past from the read voting history information. The number of past piggyback votes is the number of times that user γ has cast a vote in the past with the same predicted results as those cast by the cast members of the voting support video. The number of past piggyback votes is tallied for each cast member of the voting support video (more specifically, the reference video) based on user-related information including the voter ID and voting-related information.

[0108] The processor selects candidate videos for distribution from among multiple voting support videos based on the number of past piggyback votes identified for each performer. For example, a predetermined number of performers are selected in descending order of the number of past piggyback votes, and the voting support videos of each selected performer are selected as candidate videos for distribution. This procedure makes it possible to select candidate videos for distribution based on user γ's past piggyback voting trends (specifically, which performers they voted for most often).

[0109] In the third method, the processor reads out viewing history information from the user-related information about the user γ, whose user ID was identified in step S101, from the storage unit 34. As can be seen from FIG. 4, the viewing history information is information associated with the user ID.

[0110] The processor identifies the number of times the voting support video has been viewed in the past, i.e., the number of past views, from the read viewing information. The number of past views is tallied for each performer in the voting support video viewed in the past, based on user-related information including the voting ID and video-related information. The processor then selects candidate videos for distribution from among multiple voting support videos based on the number of past views identified for each performer. For example, a predetermined number of performers are selected in descending order of the number of past views, and voting support videos of each performer are selected as candidate videos for distribution. This procedure makes it possible to select candidate videos for distribution based on user γ's video viewing tendencies (specifically, which performer's videos the user has watched the most).

[0111] There are other possible examples of cases where criteria are set using the third method. For example, the processor may tally the past number of views for each attribute of a race based on user-related information including a betting ID, video-related information, and race-related information. In this case, the processor identifies the attribute of a race whose past number of views meets a predetermined condition, specifically, the attribute of a race whose past number of views has reached a threshold (hereinafter, referred to as the second target attribute).

[0112] The processor extracts a plurality of voting support videos related to voting in a race for a second target attribute from a plurality of voting support videos that can be distributed at that time. The processor also performs the first to third analysis processes on the extracted plurality of voting support videos in the above-described procedure, and calculates an evaluation value for each video based on the analysis results of each analysis process.

[0113] The processor then selects candidate videos to be distributed from among multiple voting support videos related to votes for races of the second target attribute, based on the evaluation value calculated for each video. This procedure makes it possible to select candidate videos to be distributed while taking into account user γ's video viewing tendencies (specifically, which race attributes the user γ has watched the most videos about). This makes it possible to select candidate videos to be distributed that reflect the attributes of races preferred by user γ.

[0114] Regarding the fourth method, the processor reads, from the storage unit 34, information about friends from the user-related information about user γ, whose user ID was identified in step S101. The information about friends is information included in the user-related information if user γ has friends (see FIG. 14). A friend is, for example, another user who has a relationship (a predetermined relationship) with user γ in which they share information about voting. When user γ and another user mutually acknowledge that they have a predetermined relationship, the two users may become associated with each other accordingly. User γ and his friends can exchange messages and the like while watching the same video using the voting app function. User γ and his friends may also be able to view each other's voting information. As can be seen from FIG. 14, friends are associated with user γ's user ID (user identification information).

[0115] The processor reads viewing history information from the user-related information of the friends of user γ, and selects a candidate video for distribution from among multiple voting support videos based on criteria set according to the read viewing history information. Specifically, for example, the processor identifies a voting support video that the friend is currently watching, and selects the same video as the voting support video or another voting support video related to a race whose result is predicted by the identified voting support video as a candidate video for distribution. Alternatively, if a friend frequently watches a certain performer's voting support video (specifically, a predetermined number of times or more within a predetermined period), the processor may identify that performer (hereinafter, the target performer).The processor may then select a video of the target performer from among the voting support videos currently being distributed as a candidate video for distribution.

[0116] As described above, when the criteria are set using the fourth method, candidate videos for distribution can be selected taking into consideration the video viewing status of user γ's friends.

[0117] Furthermore, in the fourth method, the processor may read voting history information from the user-related information of user γ's friends, and select candidate videos for distribution from among multiple voting support videos based on criteria set according to the read voting history information. Specifically, for example, the processor identifies a target race (corresponding to a target sport) from the friend's voting history information, which is a race for which the friend voted within a predetermined period. The processor then selects voting support videos associated with the target race from among the multiple voting support videos as candidate videos for distribution. Voting support videos associated with the target race are videos related to voting conducted in anticipation of the outcome of the target race, i.e., voting support videos related to the target race. According to the above procedure, candidate videos for distribution can be selected taking into account the voting status of friends, thereby making it possible to notify user γ of voting support videos for races in which friends have voted.

[0118] In the fourth method, the other users associated with the user ID of user γ are not limited to being identified as friends of user γ, and may be, for example, users who belong to the same group as user γ. Alternatively, the other users may be users designated by user γ (for example, favorite users of user γ or users followed by user γ).

[0119] [Regarding the third embodiment] In the second embodiment, candidate distribution videos are selected to reflect the preferences or requests of user γ by taking into account the filtering conditions specified by user γ. On the other hand, if machine learning is performed on the characteristics of voting support videos that user γ has previously viewed and the learning results are used to select candidate distribution videos, candidate distribution videos can be determined that more accurately reflect user γ's preferences. This embodiment (hereinafter, the third embodiment) will be described with reference to FIGS. 15 and 16. Note that the following mainly describes the differences between the third embodiment and the first embodiment. In addition, in FIG. 15, functional units that are common to the functional units of the first embodiment are assigned the same reference numerals as those in FIG. 3.

[0120] The information processing device 10X according to the third embodiment further includes a feature identification unit 39 in addition to the functional units included in the information processing device 10 according to the first embodiment. The feature identification unit 39 identifies, by learning, the features of a voted video in which the user γ voted while watching, among the voting support videos that the user γ has previously watched.

[0121] More specifically, the feature identification unit 39 identifies the features of a predetermined video from among the videos with votes through machine learning. A predetermined video is a video in which user γ, while watching the video, has piggybacked on the vote, i.e., cast a vote with the same predicted content as the vote cast by the cast of the predetermined video. The predetermined video is identified based on information associated with the user ID of user γ (specifically, viewing history information and voting history information of user-related information) and video-related information. In other words, the processor constituting the feature identification unit 39 identifies the features of the predetermined video by performing learning based on the information associated with the user ID of user γ.

[0122] Video features are the content that characterizes the video, specifically, the attributes (category) of the performers in the video, the number of performers, commonalities between members of the performer group, the performers' words and actions in the video (level of popularity), the attributes of the race in which the performers are voting, and the tendency of the performers to vote (number of votes and amount of voting). When extracting video features, known annotation tools can be applied. Annotation tools are tools that annotate the target video with tag information (metadata) related to the video.

[0123] The feature identification unit 39 performs machine learning using training data that includes a set of features (specifically, metadata) of voting support videos previously viewed by user γ and whether or not user γ has voted on those videos. The machine learning algorithm is not particularly limited, and examples that can be used include genetic programming, inductive logic programming, support vector machines, clustering, Bayesian networks, extreme learning machines (ELM), and decision tree learning. Furthermore, gradient descent or backpropagation may be used as a method for minimizing the objective function (loss function) in neural network machine learning.

[0124] In the selection process in the video distribution flow according to the third embodiment, as shown in Fig. 16, a processor of a computer constituting the information processing device 10X executes a feature identification process. In the feature identification process, the processor identifies features of a predetermined video (i.e., a video for which a user γ has previously viewed a voting support video and for which the user γ has placed a piggyback vote while viewing the video) by machine learning (S031). The feature identification process is not limited to being executed during the video distribution flow, and may be executed before the video distribution flow is implemented.

[0125] The flow of the selection process after step S031 is the same as in the first embodiment, in which the processor executes the first to third analysis processes for each of the voting support videos posted on the race day (S032).The processor then calculates an evaluation value based on the analysis results of each analysis process (S033), and selects candidate videos for distribution based on the evaluation value (S034).

[0126] Thereafter, the processor (more specifically, the list sending unit 37) sends a distribution candidate list indicating the selected distribution candidate videos to user γ. At this time, the processor determines the distribution candidate list according to the characteristics of the videos identified in step S031. Specifically, after selecting a predetermined number of distribution candidate videos in descending order of evaluation value, the processor determines whether each selected video has the characteristic identified in step S031 and adds the video having the characteristic to the distribution candidate list.

[0127] That is, in the third embodiment, the processor transmits to user γ a ​​distribution candidate list indicating distribution candidate videos that are selected based on the rating value and have predetermined video characteristics identified by machine learning. As described above, in the third embodiment, distribution candidate videos can be determined based on the preferences of user γ. In detail, videos that have the same characteristics as videos for which user γ has previously voted can be added to the distribution candidate list and recommended to user γ.

[0128] In the above case, the characteristics of the voting support videos that user γ has viewed in the past, in which user γ voted while viewing the video, are identified, but this is not limited to this. For example, the characteristics of videos with long viewing times (specifically, videos with viewing times exceeding a predetermined time) may be identified from the voting support videos that user γ has viewed in the past, and distribution candidate videos may be selected based on the identification results.

[0129] Furthermore, machine learning is not limited to being performed for the purpose of identifying the characteristics of videos viewed by user γ, but may also be performed for the purpose of identifying voting trends previously performed by user γ, regardless of whether or not the user γ has viewed a video. For example, the processor may perform machine learning based on information associated with the user ID of user γ, specifically voting history information, and set criteria for each user according to the results of the machine learning. In this case, machine learning constructs a candidate selection model as criteria for selecting distribution candidate videos.

[0130] The candidate selection model is a mathematical model that receives voting history information of user γ and outputs the suitability of each of a plurality of voting support videos as a distribution candidate video.

[0131] The voting history information that is input to the model includes, for example, information about the attributes of races for which the user has voted, the number of votes for races of each attribute, and the user's preferences for the voting content. Information about race attributes includes, for example, the type and grade of the race, the number of runners in the race, and the venue where the race is held. Information about preferences for the voting content includes, for example, information about whether the user frequently votes for voting content with high (or low) odds.

[0132] The fitness output from the model is an index value that indicates whether a voting support video is suitable as a distribution candidate video, and specifically, it is a weighting value determined according to the attributes of the race associated with each voting support video. Note that the more frequently users vote for a race attribute, the higher the fitness (weighting) of the voting support video associated with that race.

[0133] A candidate selection model is constructed for each user. Then, the processor uses the candidate selection model for each user to select, for each user, candidate videos for distribution from among multiple voting support videos. Specifically, the processor selects a predetermined number of voting support videos as candidate videos for distribution in descending order of the degree of suitability output from the model. According to the above-described procedure, distribution candidate videos can be selected taking into consideration the user's voting history, particularly the attributes of races in which the user has frequently voted.

[0134] [Regarding the fourth embodiment] The selection process may further take into account the current location of user γ, or more precisely, the location of the user terminal 12 used by user γ. In other words, a criterion may be set according to the location of the user terminal 12 associated with the user identification information, and distribution candidate videos may be selected based on that criterion. Such an embodiment (hereinafter, the fourth embodiment) will be described with reference to FIG. 17. The following description will mainly focus on the differences between the fourth embodiment and the above-described embodiments.

[0135] In the selection process according to the fourth embodiment, first, a processor of a computer constituting an information processing device communicates with the user terminal 12 of user γ and identifies the location of the user terminal 12 based on information received from the user terminal 12 (S201). There are no particular limitations on the method for identifying the location of the user terminal 12. For example, if the user terminal 12 has a location detection function using a GPS (Global Positioning System), the location of the user terminal 12 may be identified by receiving, from the user terminal 12, data indicating the location (latitude, longitude, and altitude) detected by that function.

[0136] The processor then identifies races for which betting is available at the identified location of the user terminal 12 (S202). A race for which betting is available at a certain location is, for example, a race for which betting is permitted (legal) in the place or area corresponding to that location. Hereinafter, a race for which betting is available at a certain location is referred to as a betting available race. There are no particular limitations on the method for identifying betting available races. For example, betting available races at the location of the user terminal 12 may be identified by referencing map data that shows the correspondence between each location on a map and the betting available races at each location.

[0137] After identifying the eligible races, the processor extracts multiple voting support videos related to voting for eligible races from among the multiple voting support videos currently available for distribution.The processor then performs the first to third analysis processes for each of the extracted multiple voting support videos in the above-described procedure, and calculates an evaluation value for each video based on the analysis results of each analysis process (S203, S204).The processor then selects a distribution candidate video from among the multiple voting support videos related to voting for eligible races, according to the evaluation value calculated for each video (S205).

[0138] As described above, in the fourth embodiment, candidate videos for distribution can be selected by further taking into account the location of the user terminal 12 used by user γ. This allows candidate videos for distribution to be appropriately selected, reflecting the current location of user γ.

[0139] [Other embodiments] Up to this point, the information processing device, information processing method, and program of the present invention have been described using specific examples, but the above-described embodiment is merely an example, and other embodiments may be considered.

[0140] In the above embodiment, the processor of the server computer performs the functions of the information processing device of the present invention, but some of the functions of the server computer, for example, the selection unit 36, may be realized by the user terminal 12. In other words, the processor of the information processing device of the present invention may be provided in both the server computer and the user terminal 12. In this case, the processor of the user terminal 12 may evaluate each of the multiple voting support videos based on the popularity of the performers and the number of hits, etc., and select candidate videos for distribution according to the evaluation results.

[0141] Furthermore, in the above embodiment, when evaluating each of the multiple voting support videos, each video was analyzed, specifically, the speech content or facial expressions of the performers in the video, the amount of votes cast by the performers and the number of winning votes, and the number of viewers who voted while watching the video were identified. Then, each voting support video was evaluated according to the identified analysis results (i.e., the results of the analysis performed from three perspectives), and more specifically, an evaluation value was calculated based on each of the analysis results performed from the three perspectives. However, this is not limited to this, and each voting support video may be evaluated based on any of the analysis results performed from the above three perspectives.

[0142] Furthermore, in the above embodiment, when evaluating each voting support video that can be distributed during a predetermined period, information obtained during that distribution period is used, such as the number of hits by a performer during the distribution period and the number of viewers who voted during the distribution period. In other words, in the above embodiment, even if the voting support video features the same performer, if the distribution period (distribution date) changes, the information used for the evaluation will change, and the evaluation (evaluation value) will also fluctuate. However, this is not limited to this, and voting support videos may also be evaluated using fixed information that is not affected by the distribution period (e.g., personal information of the performer, etc.).

[0143] Furthermore, in the above embodiment, the selection unit 36 ​​uses the evaluation results of each video, specifically the magnitude of the evaluation value of each video, as the criterion (predetermined criterion) when selecting candidate videos to distribute from among multiple voting support videos, but this is not limited to this. For example, without evaluating each video, candidate videos to distribute may be selected by narrowing down the search results by the user as in the second embodiment described above, or the machine learning in the third embodiment described above may be used to identify characteristics of predetermined videos, and voting support videos containing those characteristics may be selected as candidate videos to distribute.

[0144] [summary] The information processing device of the present invention is an information processing device that distributes videos related to voting to predict the results of a specified competition to users, and is equipped with a selection unit that selects some videos from a plurality of videos as candidate videos for distribution based on specified criteria, and a list transmission unit that transmits a list showing the selected candidate videos for distribution to the user. According to the above configuration, videos that can be recommended to a user based on a predetermined criterion are selected from among a plurality of videos related to voting, and the selected videos are presented to the user as candidate videos for distribution, thereby enabling effective distribution of videos to the user, so that the user can watch videos that are useful for voting.

[0145] In the information processing device of the present invention, the selection unit may select a candidate video to be distributed from among the plurality of videos, depending on an evaluation result of at least one of the audio and the image of the performers in the video. According to the above configuration, when selecting candidate videos for distribution from among multiple videos, the selection is made based on the evaluation results of the audio or video of the performers in each video, so that more appropriate videos, for example, videos in which the performers are highly popular, can be presented to the user as candidates for distribution.

[0146] In the above configuration, the selection unit may calculate an evaluation value for each of the multiple videos based on the number of performers who speak in the video, and select candidate videos for distribution based on the evaluation value of each video. In this case, it is preferable that the selection unit assign a higher evaluation value to a video in which multiple performers speak. According to the above configuration, the popularity of each video can be appropriately evaluated by calculating an evaluation value for each video based on the number of performers who speak in the video. Then, by selecting videos to be distributed according to the evaluation value, more appropriate videos (i.e., videos with a high popularity) can be presented to the user.

[0147] In the above configuration, the selection unit may identify facial expressions of performers in each of the multiple videos from images of the performers, calculate an evaluation value based on the identified facial expressions of the performers, and select candidate videos for distribution based on the evaluation value of each video. In this case, it is preferable that the selection unit assign a higher evaluation value to videos in which performers with a predetermined facial expression are shown. According to the above configuration, by calculating an evaluation value for each video based on the facial expressions of the performers in the video, it is possible to more appropriately evaluate the popularity of the video. Then, by selecting candidate videos for distribution according to the evaluation value, it is possible to present more appropriate videos (i.e., videos with a high popularity) to the user.

[0148] In the information processing device of the present invention, the selection unit may select a candidate video to be distributed from among a plurality of videos in accordance with an evaluation result based on information regarding votes by performers in the video. According to the above configuration, when selecting candidate videos for distribution from among multiple videos, the selection is made based on the evaluation results of the content of votes or the winning results by the performers of each video, so that more appropriate videos, for example, videos that are useful for the user when voting, can be presented to the user as candidates for distribution.

[0149] Furthermore, in a case where a predetermined competition is held multiple times within a predetermined period and voting is conducted for each predetermined competition, the selection unit may calculate an evaluation value for each of the multiple videos based on the numerical value of votes for which the performers of the videos correctly predicted the outcome of the predetermined competition immediately prior to the predetermined period, and select candidate videos for distribution based on the evaluation value of each video. In this case, it is preferable that the selection unit assign a higher evaluation value the larger the numerical value. According to the above configuration, for each video, an evaluation value is calculated based on the numerical value (e.g., the amount of bets or the payout) related to the votes that were correctly predicted in the specified competition in the previous round, and candidate videos for distribution are selected according to the evaluation value, so that more appropriate videos (i.e., videos that are useful for the user when voting) can be presented to the user.

[0150] The selection unit may also calculate an evaluation value for each of a plurality of videos based on the number of consecutive votes that have been correct among multiple votes cast by the performers of the videos within a predetermined period, and select candidate videos for distribution based on the evaluation value of each video. In this case, it is preferable that the selection unit assign a higher evaluation value to videos with a higher number of consecutive votes. According to the above configuration, an evaluation value is calculated for each video based on the number of consecutive votes that have been correctly predicted, and candidate videos for distribution are selected based on the evaluation value, so that more appropriate videos (i.e., videos that are useful for users when voting) can be presented to users.

[0151] In addition, in cases where a specified competition is held multiple times within a specified period and voting is conducted for each specified competition, the selection unit may calculate an evaluation value for each of the multiple videos based on the change (increase or decrease) in the amount of votes cast by the participants in the video in each vote during the specified period, and select candidate videos for distribution based on the evaluation value of each video. According to the above configuration, an evaluation value is calculated for each video based on the change (increase or decrease) in the amount of votes cast by the performers in each vote, and candidate videos for distribution are selected based on the evaluation value, so that more appropriate videos (i.e., videos that are useful for users when voting) can be presented to users.

[0152] In addition, in the information processing device of the present invention, the selection unit may select a candidate video for distribution from among a plurality of videos in accordance with an evaluation result based on information about viewers of the video who voted while watching the video. According to the above configuration, when selecting candidate videos for distribution from among multiple videos, the selection is made based on evaluation results such as the number of viewers who voted while watching each video, so that more appropriate videos, for example, videos that are useful for the user when voting, can be presented to the user as candidates for distribution.

[0153] In addition, in the above configuration, the selection unit may calculate an evaluation value for each of the multiple videos based on the number of viewers who, while watching the video, cast votes with the same predicted content as the votes cast by the video's performers, and select candidate videos for distribution based on the evaluation value of each video. According to the above configuration, for each video, an evaluation value is calculated based on the number of viewers who voted for the same predicted content as the cast of the video (i.e., piggybacking votes), and candidate videos for distribution are selected based on the evaluation value, so that more appropriate videos (i.e., videos that are useful for users when voting) can be presented to users.

[0154] In addition, in cases where the time period during which a specified competition is held and the time period during which voting is accepted alternately, the selection unit may calculate an evaluation value for each of the multiple videos during the time period during which voting is accepted, and select candidate videos for distribution based on the evaluation value of each video. According to the above configuration, for each video, an evaluation value of the video during the voting acceptance time period is calculated. Here, during the voting acceptance time period, the performers in the video are voting and the video is popular. Therefore, by selecting candidate videos for distribution based on the evaluation value of the video during the voting acceptance time period, it is possible to present users with more appropriate videos, i.e., videos with high popularity and videos that are useful for users when voting.

[0155] In addition, in cases where a specified competition is held multiple times within a specified period and voting is conducted for each specified competition, the selection unit may calculate an evaluation value for each of a plurality of videos that are distributed in real time within the specified period, each time the specified competition is held within the specified period, and select candidate videos for distribution based on the evaluation value of each video. As each predetermined competition is held within a predetermined period, the evaluation results of the videos may change accordingly. Taking this into consideration, the above configuration calculates an evaluation value for each video each time a predetermined competition is held, and distribution candidates are selected based on the evaluation value. This allows distribution candidate videos to be selected based on the current evaluation value, making it possible to effectively distribute videos that are currently recommended to users.

[0156] In addition, the information processing device of the present invention may have a feature identification unit that identifies, through learning, the features of videos with votes that the user voted on while watching, among the videos that the user has watched in the past, and the list transmission unit may transmit to the user a list showing videos that are candidates for distribution and have the features selected by the selection unit and identified by the feature identification unit. According to the above configuration, the characteristics of videos that a user has previously viewed and voted for while viewing (videos with votes) are identified, and when videos with votes are presented to the user as candidates for distribution, the characteristics of the videos with votes are reflected. This makes it possible to present videos as candidates for distribution to the user based on the user's preferences and inclinations.

[0157] In the above configuration, the feature identification unit may identify the feature of a predetermined video among the videos with votes. In this case, the predetermined video may be a video in which a user, while watching the predetermined video, cast a vote with the same predicted content as a vote by a cast member of the predetermined video. According to the above configuration, the characteristics of the voted videos in which the user has voted for the same predicted content as the cast member's vote (i.e., piggyback voting) are identified, and the characteristics are reflected when presenting candidate videos for distribution to the user. This makes it possible to present candidate videos for distribution to the user based on the user's preferences (thoughts regarding voting).

[0158] In addition, the information processing method of the present invention is an information processing method for distributing videos related to voting to predict the results of a specified competition to users, characterized in that a computer selects some videos from a plurality of videos as candidate videos for distribution based on specified criteria, and the computer transmits a list showing the selected candidate videos for distribution to the user. By using the above information processing method, videos related to voting can be effectively distributed to users, and users can watch videos that are useful for voting.

[0159] In addition, the program of the present invention is a program for distributing videos to users regarding voting to predict the results of a specified competition, and is a program that causes a computer to select some videos from a plurality of videos as candidate videos for distribution and transmit a list showing the selected candidate videos for distribution to the user. By having a computer execute the above program, videos related to voting can be effectively distributed to users, allowing users to watch videos that are useful for voting.

[0160] In addition, an information processing device according to one embodiment of the present invention is an information processing device that distributes videos relating to voting to predict the results of a specified competition to a user terminal, and is equipped with a processor, wherein the processor identifies user identification information associated with the user terminal, the processor selects some videos from a plurality of videos as candidate videos for distribution based on criteria set according to the user identification information, the processor displays a list showing the selected candidate videos for distribution on a display unit of the user terminal, and the processor displays one video selected from the candidate videos for distribution included in the list on the display unit of the user terminal. According to the above configuration, it is possible to select videos that can be recommended to a user from among multiple videos related to voting, based on criteria set according to user identification information. In other words, it is possible to select candidate videos for distribution for each user. This allows candidate videos to be selected taking into account the user's preferences and video viewing status, and presents the selected candidate videos to the user. As a result, it is possible to effectively distribute videos related to voting to users, and also allows users to watch videos that are useful for voting.

[0161] The processor may also display a video in a first display area of ​​the display unit, and display voting information associated with the video displayed in the first display area in a second display area of ​​the display unit. In this case, the processor may be able to accept votes based on the voting information displayed in the second display area through a user terminal. According to the above configuration, while a video is displayed in the first display area of ​​the display unit, voting information used by the user when voting can be displayed in the second display area of ​​the display unit, allowing the user to efficiently cast their vote based on the voting information while referring to the video information.

[0162] The processor may also select a distribution candidate video from among a plurality of videos based on a filtering condition specified through a user terminal associated with the user identification information. According to the above configuration, candidate videos for distribution are selected taking into consideration the filtering conditions specified by the user, thereby making it possible to present candidate videos for distribution to the user that reflect the user's preferences, requests, etc.

[0163] The processor may also select a candidate video to be distributed from among a plurality of videos based on criteria set according to voting history information of a user associated with the user identification information. According to the above configuration, candidate videos for distribution are selected taking into account the user's voting history, which makes it possible to present to the user candidate videos for distribution that are selected based on the attributes of the sports that the user likes or the attributes of the sports that the user's predictions have a high hit rate.

[0164] In the above configuration, the processor may identify an attribute of a predetermined sport for which a numerical value identified from the voting history information satisfies a predetermined condition. In this case, the processor may select, from among the multiple videos, a video about votes for the predetermined sport that corresponds to the identified attribute as a candidate video for distribution. According to the above configuration, it is possible to identify attributes of sports that a user likes or attributes of sports for which the user's predictions have a high success rate based on values ​​identified from the voting history information. Then, by selecting videos associated with the sports with the identified attributes as candidate videos for distribution, it is possible to more appropriately present videos that are useful to the user.

[0165] The processor may also identify, for each performer, the number of times that a user has previously voted for the same predicted content as a vote by the performer of the video based on the voting history information. In this case, the processor may select a candidate video for distribution from among the multiple videos based on the number of times identified for each performer. According to the above configuration, the number of votes for the same expected content as the vote by the performer of the video (i.e., piggyback votes) is identified for each performer of the video, and distribution candidate videos are selected based on the number identified for each performer. As a result, for example, videos of performers with a larger number of piggyback votes can be selected as distribution candidate videos. As a result, videos that are useful for the user when voting can be presented to the user.

[0166] The processor may also identify another user who has a predetermined relationship with the user and is associated with the user identification information, and in this case, the processor may select a candidate video to distribute from the plurality of videos based on criteria set according to information about the another user. According to the above configuration, it is possible to select candidate videos to be distributed by taking into consideration information about other users who have a predetermined relationship with the user. This makes it possible to present to the user, for example, videos about sports that the other users like, and further provide the user with an opportunity to interact with the other users.

[0167] In the above configuration, the processor may select a candidate video to be distributed from among a plurality of videos based on information relating to a viewing status of the video by another user. According to the above configuration, candidate videos for distribution can be selected taking into consideration the video viewing status of other users. As a result, for example, if another user frequently watches videos of a certain performer, the videos of that performer can be presented to the user as candidate videos for distribution.

[0168] In addition, in the above configuration, the processor may select a candidate video to be distributed from among the plurality of videos based on a video associated with a target sport, which is a predetermined sport for which another user has voted. According to the above configuration, candidate videos for distribution can be selected by taking into account the voting status of other users, more specifically, the attributes of the target sport for which the other users have voted. As a result, for example, videos associated with the target sport for which the other users have voted and videos associated with sports with the same attributes as the target sport can be presented to the user as candidate videos for distribution.

[0169] The processor may also select some videos from a plurality of videos as candidate videos for distribution based on criteria set by the results of machine learning using information associated with the user identification information. According to the above configuration, the system learns user preferences and trends in video viewing behavior, and uses the learning results to select candidate videos for distribution. This allows appropriate videos to be selected as candidate videos for distribution, and appropriately presents videos that are useful for voting to users.

[0170] The processor may also select a candidate video to be distributed from among a plurality of videos based on criteria set according to the location of a user terminal associated with the user identification information. According to the above configuration, it is possible to select candidate videos to be distributed taking into account the location of the user terminal used by the user. This makes it possible to select appropriate videos as candidate videos to be distributed taking into account the current location of the user.

[0171] In the above configuration, the processor may select, from among the plurality of videos, a video about voting for a predetermined sport for which voting is available at the location, as a candidate video to be distributed. According to the above configuration, when selecting candidate videos to distribute taking into consideration the user's current location, it is possible to select, for example, videos related to voting for a sport in which voting is available at the user's current location, thereby presenting the user with useful videos related to voting that can be performed at the user's current location.

[0172] In addition, when the processor selects multiple candidate videos for distribution, it may set a priority for each candidate video for distribution and cause the display unit to display a list showing each of the multiple candidate videos for distribution in order according to their priority. According to the above configuration, the user can check the candidate videos for distribution in order of priority, for example. This allows the user to efficiently check each video when the list contains multiple candidate videos for distribution.

[0173] In addition, the information processing method of the present invention is an information processing method for distributing videos related to voting to predict the results of a specified competition to a user terminal, wherein a computer identifies user identification information associated with the user terminal, the computer selects some videos from a plurality of videos as candidate videos for distribution based on criteria set according to the user identification information, the computer displays a list showing the selected candidate videos for distribution on a display unit of the user terminal, and the computer displays one video selected from the candidate videos for distribution included in the list on a display unit of the user terminal. According to the above method, distribution candidate videos can be selected for each user, and distribution candidate videos selected in consideration of the user's preferences, video viewing status, etc. can be presented to the user.

[0174] In addition, the program of the present invention is a program for distributing videos related to voting to predict the results of a specified competition to a user terminal, and causes a computer to execute the following processes: a process of identifying user identification information associated with the user terminal; a process of selecting some videos from a plurality of videos as candidate videos for distribution based on criteria set according to the user identification information; a process of displaying a list showing the selected candidate videos for distribution on the display unit of the user terminal; and a process of displaying one video selected from the candidate videos for distribution included in the list on the display unit of the user terminal. By running the above program on a computer, distribution candidate videos can be selected for each user, and distribution candidate videos selected taking into account the user's preferences, video viewing status, etc. can be presented to the user. [Explanation of symbols]

[0175] 10,10X Information Processing Device 12 User terminal 14. Telecommunications Network 21 processors 22 Memory 23 Communication Interface 24 Storage 25 Bus 31 Video Acquisition Unit 32 Vote Acquisition Department 33 Result acquisition part 34 Storage section 35 Update section 36 Selection Department 36A Video Analysis Unit 36B Evaluation value calculation unit 37 List transmission unit 38 Video Distribution Department 39 Feature Analysis Unit S Information Processing System T1 Upper area (1st display area) T2 center area (second display area) T3 Lower area (second display area)

Claims

1. An information processing device that distributes videos related to voting for predicting the results of a predetermined competition to user terminals, a processor; The processor determines a user identity associated with the user terminal; The processor selects some of the videos as distribution candidate videos from among the plurality of videos based on criteria set according to the user identification information; The processor displays a list of the selected distribution candidate videos on a display unit of the user terminal; The information processing device, wherein the processor displays one of the videos selected from the distribution candidate videos included in the list on the display unit of the user terminal.

2. the processor displays the video in a first display area of ​​the display unit, and displays voting information associated with the video displayed in the first display area in a second display area of ​​the display unit; The information processing device according to claim 1 , wherein the processor is capable of accepting, through the user terminal, a vote based on the voting information displayed in the second display area.

3. The information processing device according to claim 1 , wherein the processor selects the distribution candidate video from among the plurality of videos based on a filtering condition specified through the user terminal associated with the user identification information.

4. The information processing device according to claim 1 or 2, wherein the processor selects the distribution candidate video from among the plurality of videos based on the criteria set according to voting history information of the user associated with the user identification information.

5. The processor identifies an attribute of the predetermined competition for which a numerical value identified from the voting history information satisfies a predetermined condition; The information processing device according to claim 4 , wherein the processor selects, from among the plurality of videos, the video about a vote for the predetermined sport that corresponds to the identified attribute as the video to be distributed.

6. The processor identifies, for each performer, the number of times that the performer has previously voted for the same predicted content as the vote by the performer of the video, based on the voting history information; The information processing device according to claim 4 , wherein the processor selects the candidate videos to be distributed from among the plurality of videos based on the number of times specified for each performer.

7. The processor identifies another user who has a predetermined relationship with the user and is associated with the user identification information; The information processing device according to claim 1 , wherein the processor selects the candidate video to be distributed from among the plurality of videos based on the criteria set according to information about the other user.

8. The information processing device according to claim 7 , wherein the processor selects the candidate video to be distributed from among the plurality of videos based on information relating to a viewing status of the video by the different user.

9. The information processing device according to claim 7 , wherein the processor selects the candidate video to be distributed from among the plurality of videos based on the video associated with the target sport, which is the predetermined sport for which the other user has voted.

10. The information processing device according to claim 1 or 2, wherein the processor selects some of the videos from the plurality of videos as candidate videos for distribution based on the criteria set by the results of machine learning using information associated with the user identification information.

11. The information processing device according to claim 1 or 2, wherein the processor selects the candidate video for distribution from among the plurality of videos based on the criteria set according to the location of the user terminal associated with the user identification information.

12. The information processing device according to claim 11 , wherein the processor selects, from among the plurality of videos, the videos about voting for the predetermined sport for which voting is available at the location as the distribution candidate videos.

13. 13. The information processing device according to claim 1, wherein, when a plurality of the candidate videos for distribution are selected, the processor sets a priority for each of the candidate videos for distribution, and causes the display unit to display the list showing each of the candidate videos for distribution in an order according to the priority.

14. An information processing method for distributing, to a user terminal, a video relating to voting for predicting the outcome of a predetermined competition, comprising: a computer identifying user identification information associated with the user terminal; The computer selects some of the videos as distribution candidate videos from among the plurality of videos based on criteria set according to the user identification information; The computer displays a list of the selected distribution candidate videos on a display unit of the user terminal; An information processing method in which a computer displays one of the videos selected from the distribution candidate videos included in the list on the display unit of the user terminal.

15. A program for distributing to a user terminal a video regarding voting for predicting the outcome of a predetermined competition, identifying user identification information associated with the user terminal; a process of selecting some of the videos as distribution candidate videos from among the plurality of videos based on criteria set according to the user identification information; a process of displaying a list of the selected distribution candidate videos on a display unit of the user terminal; and displaying one of the videos selected from the distribution candidate videos included in the list on the display unit of the user terminal.

16. An information processing device that distributes to users videos relating to voting for predicting the results of a predetermined competition, a selection unit that selects some of the videos as distribution candidate videos from among the plurality of videos based on predetermined criteria; and a list sending unit that sends a list indicating the selected distribution candidate videos to a user.

17. The information processing device according to claim 16 , wherein the selection unit selects the candidate videos to be distributed from among the plurality of videos in accordance with an evaluation result regarding at least one of audio and images of performers in the videos.

18. the selection unit calculates an evaluation value for each of the plurality of videos based on the number of performers who speak in the video, and selects the candidate videos for distribution according to the evaluation value of each of the videos; The information processing device according to claim 17 , wherein the selection unit increases the evaluation value of the video in which a plurality of performers are speaking.

19. the selection unit identifies facial expressions of performers in each of the plurality of videos from images of the performers, calculates an evaluation value based on the identified facial expressions of the performers, and selects the candidate videos for distribution according to the evaluation value of each of the videos; The information processing device according to claim 17 or 18, wherein the selection unit increases the evaluation value of the video in which a performer with a predetermined facial expression appears.

20. The information processing device according to claim 16 , wherein the selection unit selects the distribution candidate video from among the plurality of videos in accordance with an evaluation result based on information regarding votes by performers for the video.

21. In the case where the predetermined competition is held multiple times within a predetermined period and voting is conducted for each predetermined competition, the selection unit calculates an evaluation value for each of the plurality of videos based on a numerical value related to votes for which the performers of the videos correctly predicted the predetermined competition in the most recent event during the predetermined period, and selects the distribution candidate videos in accordance with the evaluation value of each of the videos; The information processing device according to claim 20 , wherein the selection unit sets the evaluation value to a higher value as the numerical value increases.

22. the selection unit calculates an evaluation value for each of the plurality of videos based on the number of consecutive votes in which the prediction was correct among multiple votes cast by the cast members of the videos within a predetermined period, and selects the distribution candidate videos according to the evaluation value of each of the videos; The information processing device according to claim 20 , wherein the selection unit sets the evaluation value to a higher value as the number of consecutive occurrences increases.

23. In the case where the predetermined competition is held multiple times within a predetermined period and voting is conducted for each predetermined competition, 23. The information processing device according to claim 20, wherein the selection unit calculates an evaluation value for each of the plurality of videos based on the change in the amount of votes cast in each vote conducted by the cast members of the video during the specified period, and selects the candidate videos for distribution based on the evaluation value of each of the videos.

24. The information processing device according to claim 16 , wherein the selection unit selects the candidate video for distribution from among the plurality of videos in accordance with an evaluation result based on information about viewers of the video who voted while watching the video.

25. The information processing device according to claim 24, wherein the selection unit calculates an evaluation value for each of the plurality of videos based on the number of viewers who, while watching the video, cast votes with the same predicted content as those cast by the performers of the video, and selects the candidate videos for distribution based on the evaluation value of each of the videos.

26. 26. An information processing device as claimed in any one of claims 16 to 25, wherein, in a case where the time period during which the specified competition is held and the time period during which voting is accepted alternately, the selection unit calculates an evaluation value for each of the plurality of videos during the time period during which voting is accepted, and selects the candidate videos for distribution based on the evaluation value of each of the videos.

27. In the case where the predetermined competition is held multiple times within a predetermined period and voting is conducted for each predetermined competition, 27. An information processing device as claimed in any one of claims 16 to 26, wherein the selection unit calculates an evaluation value for each of the plurality of videos distributed in real time within the specified period each time the specified competition is held within the specified period, and selects the candidate videos for distribution based on the evaluation value of each of the videos.

28. a feature identification unit that identifies, by learning, the features of a voted video on which the user voted while viewing, among the videos that the user has previously viewed; 28. The information processing device according to claim 16, wherein the list sending unit sends, to a user, a list indicating the distribution candidate videos selected by the selection unit and having the characteristics identified by the characteristic identifying unit.

29. The feature identification unit identifies the feature of a predetermined video among the videos with votes, The information processing device according to claim 28 , wherein the predetermined video is a video for which the user, while watching the predetermined video, has cast a vote with the same predicted content as a vote by a performer of the predetermined video.

30. An information processing method for distributing to users a video regarding voting for predicting the outcome of a predetermined competition, comprising: The computer selects some of the videos as distribution candidate videos from among the plurality of videos based on predetermined criteria; An information processing method in which a computer transmits a list indicating the selected distribution candidate videos to a user.

31. A program for distributing to users videos relating to voting for predicting the outcome of a predetermined competition, On the computer, Selecting some of the videos as candidate videos for distribution from among the plurality of videos; A program that causes a list indicating the selected distribution candidate videos to be transmitted to a user.

Images