Information processing device, information processing method, and computer program

Abstract

To evaluate an indirect attack side player in a team sport.SOLUTION: An information processing device for evaluating attack side players in a team sport for determining winning / losing according to the amount of specific events includes an actual position specification part, an average position prediction part, a first evaluation part, and a second evaluation part. The actual position specification part specifies the actual positions of a direct attack side player who directly brings about a specific event, indirect attack side players, and defense side players at a first time and a second time. The average position prediction part predicts average positions at the second time with the indirect attach side players and the defense side players as objects. The first evaluation part calculates an actual evaluation value with the direct attack side player as an object on the basis of the actual position at the second time, and calculates a virtual evaluation value on the basis of the actual position of the direct attack side player, and the average positions of the indirection attack side players and the defense side players. The second evaluation part calculates evaluation values of the indirect attack side players on the basis of difference between the actual evaluation values and the virtual evaluation values.SELECTED DRAWING: Figure 4

Description

[Technical field]

[0001] The technology disclosed in this specification relates to an information processing device or the like for evaluating an offensive player in a team sport in which victory or defeat is determined by the frequency of a specific event. [Background technology]

[0002] In recent years, with the development of measurement technology, it has become possible to use data on the movements of each player at each moment during a match in team sports such as soccer, where the winner is decided by the number of goals (points) scored between teams of multiple players, and it is hoped that this data can be used to evaluate each player.

[0003] A method called Valuing Actions by Estimating Probabilities (hereinafter referred to as "VAEP") has been proposed as a method for evaluating players in such team sports, which evaluates future actions that are likely to score / least likely to concede a goal based on score predictions (see Non-Patent Document 1). In addition, a probabilistic rule-based model called Off-Ball Scoring Oppotunities (hereinafter referred to as "OBSO") has also been proposed as an evaluation method for plays that may lead to a goal, which evaluates the quality of positioning to receive the ball before a shot (see Non-Patent Document 2). [Prior art documents] [Non-patent literature]

[0004] [Non-Patent Document 1] Tom Decroos and 3 others, "Actions Speak Louder than Goals: Valuing Player Actions in Soccer," [online], August 2019, KDD'19, [Retrieved July 1, 2021], Internet<URL:https: / / arxiv.org / abs / 1802.07127> [Non-Patent Document 2] William Spearman, "Beyond Expected Goals", [online], 2018, MIT Sloan Conference on Sports Analytics, [Retrieved July 1, 2021], Internet<URL:https: / / www.researchgate.net / publication / 327139841_Beyond_Expected_Goals> Summary of the Invention [Problem to be solved by the invention]

[0005] The above-mentioned VAEP method predicts a score (goal), which is a rare event in the entire game, so it is not possible to obtain a stable evaluation, and it is difficult to evaluate various plays leading up to the score and loss. In addition, the above-mentioned OBSO model is a model that represents the possibility that a player who does not currently have the ball will score, so it is difficult to evaluate other offensive players who do not have the ball and do not receive the ball but move strategically to increase the team's chances of scoring with a method using this model. In this way, while the conventional player evaluation method for team sports can objectively evaluate an offensive player who directly causes a score (goal) (hereinafter referred to as a "direct offensive player"), it has a problem that it is difficult to objectively evaluate other offensive players who do not directly cause a score but move strategically to increase the team's chances of scoring (hereinafter referred to as an "indirect offensive player").

[0006] This issue is not limited to the number of goals (points scored), but is a common issue when evaluating offensive players in team sports where victory or defeat is decided by the number of specific events.

[0007] This specification discloses a technique that can solve the above-mentioned problems. [Means for solving the problem]

[0008] The technology disclosed in this specification can be realized, for example, in the following forms.

[0009] (1) The information processing device disclosed in this specification is a device for evaluating an offensive player in a team sport in which victory or defeat is decided by the amount of a specific event. The information processing device includes an actual position identification unit, an average position prediction unit, a first evaluation unit, and a second evaluation unit. The actual position identification unit identifies the actual positions of a direct offensive player who is an offensive player who directly causes the specific event, an indirect offensive player who is an offensive player other than the direct offensive player, and a defensive player at a first time point during the team sport and a second time point that is a predetermined time after the first time point. The average position prediction unit predicts the average positions of the indirect offensive player and the defensive player at the second time point using the actual positions at the first time point and a predetermined trajectory prediction model. The first evaluation unit uses a predetermined evaluation method for the direct offensive player to calculate an actual evaluation value based on the actual positions of the direct offensive player, the indirect offensive player, and the defensive player at the second time, and calculates a virtual evaluation value based on the actual position of the direct offensive player and the average position of the indirect offensive player and the defensive player at the second time. The second evaluation unit calculates an indirect evaluation value, which is the evaluation value of the indirect offensive player, based on the difference between the actual evaluation value and the virtual evaluation value of the direct offensive player.

[0010] In this way, the information processing device calculates an indirect evaluation value for an indirect offensive player based on the difference between the actual evaluation value and the virtual evaluation value of a direct offensive player who directly causes a specific event that affects the team's victory or defeat, thereby making it possible to quantitatively evaluate the movements of an indirect offensive player that contribute to an increase or decrease in the evaluation value for a direct offensive player, that is, that contribute to an improvement in the probability of a specific event occurring, and thus to an improvement in the probability of victory. Thus, the information processing device makes it possible to realize an objective evaluation of an indirect offensive player who does not directly cause a specific event that affects the team's victory or defeat, but who takes strategic actions to help the team win.

[0011] (2) In the above information processing device, the first evaluation unit may be configured to correct the actual evaluation value of the direct offensive player by subtracting the indirect evaluation value of the indirect offensive player from the actual evaluation value of the direct offensive player. This information processing device can quantify the actual evaluation of the direct offensive player after subtracting the contribution of the indirect offensive player.

[0012] (3) In the above information processing device, the specific event may be a goal in which the ball reaches a specific target area. According to this information processing device, although the player does not directly cause a goal, it is possible to objectively evaluate the probability of a goal being scored, and in turn, an indirect attacking player who takes a strategic action to help the team win.

[0013] The technology disclosed in this specification can be realized in various forms, such as an information processing device, an information processing method, a computer program for realizing those methods, a non-transitory recording medium on which that computer program is recorded, etc. [Brief description of the drawings]

[0014] [Figure 1] FIG. 1 is a block diagram showing a schematic configuration of an information processing device 100 according to an embodiment of the present invention. [Diagram 2]A flowchart showing the offensive player evaluation process according to the present embodiment. [Diagram 3] FIG. 13 is an explanatory diagram conceptually illustrating the offensive player evaluation process according to the present embodiment; [Figure 4] FIG. 13 is an explanatory diagram conceptually illustrating the offensive player evaluation process according to the present embodiment; [Diagram 5] FIG. 13 is an explanatory diagram conceptually illustrating the offensive player evaluation process according to the present embodiment; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] A. Embodiment: A-1. Configuration of information processing device 100: FIG. 1 is a block diagram showing a schematic configuration of an information processing device 100 in this embodiment. The information processing device 100 in this embodiment is a device for evaluating offensive players in team sports in which victory or defeat is decided by the number of specific events. More specifically, the information processing device 100 in this embodiment is a device for evaluating offensive players in soccer in which victory or defeat is decided by the number of goals (scores) as a specific event. In team sports such as soccer, in addition to players who are holding the ball and about to shoot, and players who are not holding the ball but are receiving the ball and about to shoot, that is, offensive players who directly cause goals (scores) (direct offensive players), offensive players who are not holding the ball and are not receiving the ball (indirect offensive players) can also increase the team's scoring probability by moving strategically, and ultimately contribute to the team's victory. The information processing device 100 in this embodiment can realize objective evaluation of indirect offensive players in addition to objective evaluation of direct offensive players.

[0016] In this specification, an offensive player is a player belonging to a team (offensive team) that attempts to cause the above-mentioned specific event to occur at each time during a match of a group sport in which teams consisting of multiple players compete, and in a ball game such as soccer, for example, it means a player belonging to a team that has possession of the ball. Moreover, a defensive player is a player belonging to a team (defensive team) other than the offensive team. During a match of a group sport, the offensive team and the defensive team switch places depending on the situation.

[0017] The information processing device 100 is configured with a computer such as a personal computer, a server, a tablet terminal, a smartphone, etc. The information processing device 100 includes a control unit 110, a storage unit 130, a display unit 152, an operation input unit 156, and an interface unit 158. These units are connected to each other via a bus 190 so as to be able to communicate with each other.

[0018] The display unit 152 of the information processing device 100 is configured, for example, by a liquid crystal display or an organic EL display, and displays various images and information. The operation input unit 156 is configured, for example, by a keyboard, a mouse, buttons, a microphone, and the like, and receives operations and instructions from an administrator. The display unit 152 may be provided with a touch panel to function as the operation input unit 156. The interface unit 158 ​​is configured, for example, by a LAN interface or a USB interface, and communicates with other devices via wired or wireless communication.

[0019] The storage unit 130 of the information processing device 100 is composed of, for example, a ROM, a RAM, a hard disk drive (HDD), etc., and is used to store various programs and data, and as a work area when various programs are executed, and as a temporary storage area for data. For example, the storage unit 130 stores an offensive player evaluation program CP, which is a computer program for executing an offensive player evaluation process described below. The offensive player evaluation program CP is provided in a state stored in a computer-readable recording medium (not shown), such as a CD-ROM, DVD-ROM, or USB memory, and is stored in the storage unit 130 by installing it in the information processing device 100.

[0020] Furthermore, in advance or during the offensive player evaluation process described below, a trajectory prediction model MO, game situation information GSI, average position information API, direct evaluation information DVI, and indirect evaluation information IVI are stored in the storage unit 130 of the information processing device 100. The contents of these will be described in conjunction with the offensive player evaluation process described below.

[0021] The control unit 110 of the information processing device 100 is configured with, for example, a CPU, and controls the operation of the information processing device 100 by executing a computer program read from the storage unit 130. For example, the control unit 110 functions as an offensive player evaluation processing unit 120 for executing an offensive player evaluation process described below, by reading an offensive player evaluation program CP from the storage unit 130 and executing it. The offensive player evaluation processing unit 120 includes a game situation information acquisition unit 121, an actual position identification unit 122, an average position prediction unit 123, a first evaluation unit 124, and a second evaluation unit 125. The functions of each of these units will be described in conjunction with the description of the offensive player evaluation process described below.

[0022] A-2. Offensive player evaluation process: Next, an offensive player evaluation process executed by the information processing device 100 of this embodiment will be described. Fig. 2 is a flowchart showing the offensive player evaluation process in this embodiment. Figs. 3 to 5 are explanatory diagrams conceptually showing the offensive player evaluation process in this embodiment.

[0023] The offensive player evaluation process is a process for objectively evaluating direct offensive players (players who are holding the ball and about to shoot, and players who are not holding the ball but are receiving the ball and about to shoot) who directly cause goals (scores) and indirect offensive players (players who are not holding the ball and do not receive the ball but move strategically to increase the team's chances of scoring) in a specific situation (hereinafter referred to as the "score target situation") in a specific soccer match (hereinafter referred to as the "match to be evaluated"). As is well known, a goal as a specific event that determines victory or defeat in soccer is a state in which the ball b reaches the physical goal G (see FIG. 3, etc.) of the opposing team, and one point is awarded to the team that caused the goal. The goal G is an example of a target area in the claims.

[0024] The offensive player evaluation process is started, for example, in response to an administrator inputting an instruction to start the process via the operation input unit 156 of the information processing device 100. First, the game situation information acquisition unit 121 (FIG. 1) of the information processing device 100 acquires game situation information GSI (S110). The game situation information GSI is data indicating the position of each player and the position of the ball at each time during the evaluation target match. The game situation information GSI is generated, for example, by satellite positioning or by analysis of images captured by multiple cameras, and is provided, for example, via a network such as the Internet. The game situation information acquisition unit 121 acquires the game situation information GSI via the interface unit 158, and stores the acquired game situation information GSI in the storage unit 130. Note that the game situation information acquisition unit 121 may acquire game situation information GSI for the entire evaluation target match, or may acquire only game situation information GSI for the periphery of the evaluation target situation of the evaluation target match.

[0025] The game situation information GSI makes it possible to grasp the game situation S at each time during the evaluation target match. FIG. 3 shows an example of the game situation S(t1) at a first time t1 of the evaluation target match, that is, the actual positions Poa(t1), Pob(t1) of the offensive players Oa, Ob, the actual positions Pda(t1), Pdb(t1) of the defensive players Da, Db, and the positions of the ball b and the goalkeeper Dk. Similarly, FIG. 4 shows the game situation S(t2) at a second time t2 after a predetermined time (e.g., 5 seconds) has elapsed from the first time t1, that is, the actual positions Poa(t2), Pob(t2) of the offensive players Oa, Ob, the actual positions Pda(t2), Pdb(t2) of the defensive players Da, Db, and the positions of the ball b and the goalkeeper Dk. In addition, in FIG. 4, the actual trajectory AT followed by each player between the first time t1 and the second time t2 is shown by a dashed line.

[0026] Referring to the game situation S(t1) at a first time t1 shown in Fig. 3 and the game situation S(t2) at a second time t2 shown in Fig. 4, in this scene, one offensive player Oa is holding the ball b and is about to shoot, while another offensive player Ob is not holding the ball b and does not receive the ball b, but is making a move to attract the defensive players Da and Db to free up the offensive player Oa in order to increase the scoring probability of the offensive player Oa. That is, in this scene, the offensive player Oa corresponds to a direct offensive player, and the offensive player Ob corresponds to an indirect offensive player.

[0027] Next, the actual position identification unit 122 (FIG. 1) of the information processing device 100 refers to the game situation information GSI to identify the actual positions P of each player at a first time t1 and a second time t2 during the match to be evaluated (S120). Here, the second time t2 is the time of a scene selected as the evaluation target situation in the match to be evaluated. The evaluation target situation may be selected arbitrarily by the administrator, may be automatically selected by the offensive player evaluation processing unit 120 of the information processing device 100 using a predetermined algorithm, or may be selected by the administrator from candidates listed by the offensive player evaluation processing unit 120.

[0028] In S120, the actual position identification unit 122 identifies the actual position P of the evaluation target player. In this embodiment, the evaluation target players are the direct offensive player, and the indirect offensive players and defensive players located in the surrounding area of ​​the direct offensive player (for example, an area within a predetermined distance from the direct offensive player). In the following description, the evaluation target players are the four players (offensive players Oa, Ob and defensive players Da, Db) shown in Figures 3 and 4.

[0029] Next, the average position prediction unit 123 (FIG. 1) of the information processing device 100 predicts the average position P' at the second time t2 for the indirect attacking player Ob and the defending players Da, Db among the evaluation target players, using the actual position P of each player at the first time t1 and the trajectory prediction model MO (S130). The storage unit 130 stores a trajectory prediction model MO acquired in advance. The trajectory prediction model MO is a known model created, for example, by a predetermined machine learning, and is a model for predicting the trajectory of the average (standard) movement of each player during a soccer game. As such a model, for example, the model described in "Trajectory Generation by Imitation Learning Considering the Evaluation of the Defensive Team in Soccer" by Masashi Teranishi and two others, 34th National Conference of the Japanese Society for Artificial Intelligence in 2020, can be used. The average position prediction unit 123 predicts the average position P' of the indirect offensive player Ob and the defensive players Da, Db at the second time t2 by inputting the actual position P of each player at the first time t1 into the trajectory prediction model MO. The average position P' indicates a position where an average player is likely to be located at the second time t2 after a predetermined time has elapsed when the average player is placed in the game situation S(t1) at the first time t1 shown in FIG. 3. Information indicating the prediction result of the average position P' is stored in the storage unit 130 as average position information API.

[0030] The average position prediction unit 123 predicts the average position P' of the indirect offensive player Ob and the defensive players Da and Db at the second time t2, and thus it is possible to assume a virtual game situation S'(t2) in which the indirect offensive player Ob and the defensive players Da and Db at the second time t2 are replaced from their actual positions P to their average positions P'. FIG. 5 shows an example of the virtual game situation S'(t2) at the second time t2 of the evaluation target match. That is, FIG. 5 shows the actual position Poa(t2) of the direct offensive player Oa, the average position P'ob(t2) of the indirect offensive player Ob, and the average positions P'da(t2) and P'db(t2) of the defensive players Da and Db at the second time t2. FIG. 5 also shows a predicted trajectory PT predicted to be followed by an average player between the first time t1 and the second time t2 by a dashed line. In the example shown in Figure 5, the indirect attacking player Ob is moving toward the direct attacking player Oa who is holding the ball b, resulting in a situation in which the defending players Da and Db are standing in front of the direct attacking player Oa.

[0031] Next, the first evaluation unit 124 (FIG. 1) of the information processing device 100 calculates an evaluation value for the direct offensive player Oa using a predetermined evaluation method (S140). As the predetermined evaluation method, any known method can be adopted, for example, the above-mentioned method called VAEP or a method using a model called OBSO can be adopted. The first evaluation unit 124 calculates an evaluation value (hereinafter referred to as "actual evaluation value Voa") in the actual game situation S(t2) based on the actual positions P of the direct offensive player Oa, the indirect offensive player Ob, and the defensive players Da and Db at the second time t2, and calculates an evaluation value (hereinafter referred to as "virtual evaluation value V'oa") in the virtual game situation S'(t2) based on the actual position P of the direct offensive player Oa and the average position P' of the indirect offensive player Ob and the defensive players Da and Db at the second time t2. The calculation result of the evaluation value for the direct attacking player Oa is stored in the storage unit 130 as direct evaluation information DVI indicating the calculation result.

[0032] FIG. 4 shows how the actual evaluation value Voa for the direct attacking player Oa is calculated, and FIG. 5 shows how the virtual evaluation value V'oa for the direct attacking player Oa is calculated. As described above, in the actual game situation S(t2) shown in FIG. 4, the movement of the indirect attacking player Ob attracts the defending players Da and Db, leaving the direct attacking player Oa free, so the actual evaluation value Voa for the direct attacking player Oa is considered to be relatively high. On the other hand, in the virtual game situation S'(t2) shown in FIG. 5, the indirect attacking player Ob moves closer to the direct attacking player Oa who is holding the ball b, so that the defending players Da and Db stand in front of the direct attacking player Oa, and the virtual evaluation value V'oa for the direct attacking player Oa is considered to be lower than the actual evaluation value Voa. In other words, it can be understood that the actual movement of the indirect offensive player Ob has improved the evaluation value for the direct offensive player Oa from the average value.

[0033] Next, the second evaluation unit 125 (FIG. 1) of the information processing device 100 calculates an evaluation value (hereinafter referred to as an "indirect evaluation value Vob") for the indirect attacking player Ob based on the difference between the actual evaluation value Voa and the virtual evaluation value V'oa for the direct attacking player Oa (S150). In this way, by calculating the indirect evaluation value Vob for the indirect attacking player Ob based on the difference between the actual evaluation value Voa and the virtual evaluation value V'oa, it is possible to quantitatively evaluate the movement of the indirect attacking player Ob that contributes to an increase or decrease in the evaluation value for the direct attacking player Oa. The calculation result of the evaluation value for the indirect attacking player Ob is stored in the storage unit 130 as indirect evaluation information IVI indicating the result.

[0034] In this embodiment, the difference between the actual evaluation value Voa and the virtual evaluation value V'oa for the direct offensive player Oa is calculated as the indirect evaluation value Vob for the indirect offensive player Ob. That is, in this embodiment, Vob=Voa-V'oa. FIG. 4 shows how the indirect evaluation value Vob for the indirect offensive player Ob is calculated. As described above, in the examples shown in FIG. 3 to FIG. 5, since the actual evaluation value Voa for the direct offensive player Oa is considered to be higher than the virtual evaluation value V'oa, the indirect evaluation value Vob for the indirect offensive player Ob is considered to be a positive value. On the other hand, if the actual evaluation value Voa for the direct offensive player Oa is lower than the virtual evaluation value V'oa, the indirect evaluation value Vob for the indirect offensive player Ob is a negative value. In the examples shown in Figures 3 to 5, there is only one indirect offensive player Ob to be evaluated, but if there are multiple indirect offensive players Ob to be evaluated, an indirect evaluation value Vob is calculated for each indirect offensive player Ob.

[0035] Next, the first evaluation unit 124 (FIG. 1) of the information processing device 100 corrects the actual evaluation value Voa for the direct attacking player Oa by subtracting the indirect evaluation value Vob for the indirect attacking player Ob from the actual evaluation value Voa for the direct attacking player Oa (S160). By performing such a correction, it is possible to realize quantification of the actual evaluation of the direct attacking player Oa minus the contribution of the indirect attacking player Ob. As described above, in the examples shown in FIGS. 3 to 5, the indirect evaluation value Vob for the indirect attacking player Ob is considered to be a positive value, so that the actual evaluation value Voa for the direct attacking player Oa is reduced by the correction. In the examples shown in Figures 3 to 5, there is only one indirect offensive player Ob to be evaluated, but if there are multiple indirect offensive players Ob to be evaluated, the sum of the indirect evaluation values ​​Vob calculated for each indirect offensive player Ob is subtracted from the actual evaluation value Voa for the direct offensive player Oa.

[0036] A-3. Advantages of this embodiment: As described above, the information processing device 100 of the present embodiment is a device for evaluating offensive players in team sports in which victory or defeat is decided by the number of goals (points) scored as a specific event. The information processing device 100 includes an actual position identification unit 122, an average position prediction unit 123, a first evaluation unit 124, and a second evaluation unit 125. The actual position identification unit 122 identifies actual positions P of a direct offensive player Oa who is an offensive player who directly causes a goal, an indirect offensive player Ob who is an offensive player other than the direct offensive player Oa, and defensive players Da, Db at a first time t1 during the team sports and a second time t2 a predetermined time after the first time t1. The average position prediction unit 123 predicts an average position P' at the second time t2 for the indirect offensive player Ob and the defensive players Da, Db using the actual positions P at the first time t1 and a predetermined trajectory prediction model MO. The first evaluation unit 124 uses a predetermined evaluation method for the direct offensive player Oa to calculate an actual evaluation value Voa based on the actual positions P of the direct offensive player Oa, the indirect offensive player Ob, and the defensive players Da, Db at the second time t2, and calculates a virtual evaluation value V'oa based on the actual position P of the direct offensive player Oa and the average position P' of the indirect offensive player Ob and the defensive players Da, Db at the second time t2. The second evaluation unit 125 calculates an indirect evaluation value Vob, which is the evaluation value of the indirect offensive player Ob, based on the difference between the actual evaluation value Voa of the direct offensive player Oa and the virtual evaluation value V'oa.

[0037] In this manner, the information processing device 100 of this embodiment calculates the indirect evaluation value Vob for the indirect offensive player Ob based on the difference between the actual evaluation value Voa and the virtual evaluation value V'oa of the direct offensive player Oa who directly causes a goal as a specific event that affects the victory or defeat of the team, thereby making it possible to quantitatively evaluate the movement of the indirect offensive player Ob that contributes to an increase or decrease in the evaluation value for the direct offensive player Oa, that is, that contributes to an improvement in the team's scoring probability and, in turn, to an improvement in the winning probability. Thus, the information processing device 100 of this embodiment makes it possible to realize an objective evaluation of the indirect offensive player who does not directly cause a goal but moves strategically to help the team win.

[0038] Furthermore, in the information processing device 100 of this embodiment, the first evaluation unit 124 corrects the actual evaluation value Voa of the direct attacking player Oa by subtracting the indirect evaluation value Vob of the indirect attacking player Ob from the actual evaluation value Voa of the direct attacking player Oa. Therefore, according to the information processing device 100 of this embodiment, it is possible to realize quantification of the substantial evaluation of the direct attacking player Oa by subtracting the contribution of the indirect attacking player Ob.

[0039] B. Variations: The technology disclosed in this specification is not limited to the above-described embodiments, and can be modified in various forms without departing from the spirit of the invention. For example, the following modifications are also possible.

[0040] The configuration of the information processing device 100 in the above embodiment is merely an example and can be modified in various ways. For example, the information processing device 100 may include an imaging device such as a camera. In this case, the information processing device 100 may directly acquire the game situation information GSI by imaging the evaluation target match with the imaging device.

[0041] In addition, in the above-described embodiment, a part of the configuration realized by hardware may be replaced by software, and conversely, a part of the configuration realized by software may be replaced by hardware.

[0042] In addition, the contents of the offensive player evaluation process in the above embodiment are merely an example and can be modified in various ways. For example, in the above embodiment, the difference between the actual evaluation value Voa and the virtual evaluation value V'oa for the direct offensive player Oa is calculated as the indirect evaluation value Vob for the indirect offensive player Ob, but the indirect evaluation value Vob may be calculated by performing some kind of calculation process on the difference (for example, multiplication by a predetermined coefficient) instead of using the difference itself.

[0043] In the above embodiment, the actual evaluation value Voa of the direct attacking player Oa is corrected by subtracting the indirect evaluation value Vob of the indirect attacking player Ob from the actual evaluation value Voa of the direct attacking player Oa, but the actual evaluation value Voa of the direct attacking player Oa may be corrected by subtracting a value obtained by performing some calculation process (for example, multiplication of a predetermined coefficient) on the indirect evaluation value Vob instead of the indirect evaluation value Vob itself. Also, the correction of the actual evaluation value Voa of the direct attacking player Oa may not be executed.

[0044] In addition, in the above embodiment, a method called VAEP and a method using a model called OBSO are used to calculate the actual evaluation value Voa and the virtual evaluation value V'oa for the direct attacking player Oa, but any other evaluation method may be used to calculate these evaluation values.

[0045] In the above embodiment, the evaluation of each offensive player in one evaluation target situation in an evaluation target match has been described, but evaluation may be performed in the same manner for multiple evaluation target situations in an evaluation target match, and an overall evaluation of each offensive player may be performed based on each evaluation result. Alternatively, evaluation may be performed in the same manner for multiple evaluation target situations in multiple evaluation target matches, and an overall evaluation of each offensive player may be performed based on each evaluation result.

[0046] In the above embodiment, the players to be evaluated are the direct attacking player and the indirect attacking players and defensive players who are located in the surrounding area of ​​the direct attacking player (for example, an area within a predetermined distance from the direct attacking player), but the method of setting the players to be evaluated may be other methods. For example, the players to be evaluated may be the direct attacking player and the indirect attacking players and defensive players who are located in a pre-defined area (for example, the so-called attacking third) close to the goal G (target area) within the playing area of ​​the team sport.

[0047] In the above embodiment, an application example to soccer, which is a group sport in which victory or defeat is decided by the number of goals (points), has been described, but the technology disclosed in this specification can be similarly applied to other group sports in which victory or defeat is decided by the number of goals. Examples of such group sports include basketball, rugby, American football, hockey, handball, and lacrosse. The technology disclosed in this specification can also be similarly applied to other group sports in which victory or defeat is decided not only by goals but also by the number of specific events. Examples of such other group sports include games such as tag in which players are divided into an escape team (defensive team) and a capture team (attacking team), and the winner is decided by the number of events in which each player of the capture team (attacking players) captures each player of the escape team (defensive players), and people moving with a purpose at an event venue. [Explanation of symbols]

[0048] 100: Information processing device 110: Control unit 120: Offensive player evaluation processing unit 121: Game situation information acquisition unit 122: Actual position identification unit 123: Average position prediction unit 124: First evaluation unit 125: Second evaluation unit 130: Memory unit 152: Display unit 156: Operation input unit 158: Interface unit 190: Bus API: Average position information AT: Actual trajectory CP: Offensive player evaluation program DVI: Direct evaluation information GSI: Game situation information IVI: Indirect evaluation information MO: Trajectory prediction model PT: Predicted trajectory

Claims

1. An information processing device for evaluating an offensive player in a team sport in which victory or defeat is decided based on the number of specific events, an actual position identification unit that identifies actual positions of a direct offensive player who is the offensive player who directly causes the specific event, an indirect offensive player who is the offensive player other than the direct offensive player, and a defensive player at a first time point and a second time point that is a predetermined time after the first time point during the team sport; an average position prediction unit that predicts average positions of the indirect offensive player and the defensive player at the second time using the actual positions at the first time and a predetermined trajectory prediction model; a first evaluation unit that uses a predetermined evaluation method to calculate an actual evaluation value for the direct offensive player based on the actual positions of the direct offensive player, the indirect offensive player, and the defensive player at the second time, and calculates a virtual evaluation value based on the actual position of the direct offensive player and the average position of the indirect offensive player and the defensive player at the second time; a second evaluation unit that calculates an indirect evaluation value, which is an evaluation value of the indirect offensive player, based on a difference between the actual evaluation value and the virtual evaluation value of the direct offensive player; Equipped with Information processing device.

2. 2. The information processing device according to claim 1, The first evaluation unit corrects the actual evaluation value of the direct offensive player by subtracting the indirect evaluation value of the indirect offensive player from the actual evaluation value of the direct offensive player. Information processing device.

3. 3. The information processing device according to claim 1, The particular event is a goal in which the ball arrives within a particular target area. Information processing device.

4. An information processing method for evaluating an offensive player in a team sport in which victory or defeat is determined by the frequency of a specific event, comprising: identifying actual positions of a direct offensive player who is the offensive player who directly causes the specific event, an indirect offensive player who is the offensive player other than the direct offensive player, and a defensive player at a first time point and a second time point that is a predetermined time after the first time point during the team sport; predicting average positions of the indirect offensive player and the defensive player at the second time using the actual positions at the first time and a predetermined trajectory prediction model; a step of calculating an actual evaluation value based on the actual positions of the direct offensive player, the indirect offensive player, and the defensive player at the second time using a predetermined evaluation method, and calculating a virtual evaluation value based on the actual position of the direct offensive player and the average position of the indirect offensive player and the defensive player at the second time; calculating an indirect evaluation value, which is an evaluation value of the indirect offensive player, based on a difference between the actual evaluation value and the virtual evaluation value of the direct offensive player; Equipped with Information processing methods.

5. A computer program for evaluating an offensive player in a team sport in which victory or defeat is determined by the number of specific events, On the computer, a process for identifying actual positions of a direct offensive player who is the offensive player who directly causes the specific event, an indirect offensive player who is the offensive player other than the direct offensive player, and a defensive player at a first time point and a second time point that is a predetermined time after the first time point during the team sport; predicting average positions of the indirect offensive player and the defensive player at the second time using the actual positions at the first time and a predetermined trajectory prediction model; a process of calculating an actual evaluation value based on the actual positions of the direct offensive player, the indirect offensive player, and the defensive player at the second time using a predetermined evaluation method, and calculating a virtual evaluation value based on the actual position of the direct offensive player and the average position of the indirect offensive player and the defensive player at the second time; calculating an indirect evaluation value, which is an evaluation value of the indirect offensive player, based on a difference between the actual evaluation value and the virtual evaluation value of the direct offensive player; Execute the Computer program.

Images