FRAUD ESTIMATION DEVICE, FRAUD ESTIMATION METHOD, AND PROGRAM

Abstract

Provided are a fraud estimation device and the like capable of reducing manual work.　A fraud estimation device 10 according to one aspect of the present disclosure comprises: a fraud estimation unit 130 that estimates the degree of fraud possibility of a participant in a subject race from estimation information including a race state information extracted from measurement data on the subject race, using an estimation model generated to estimate the degree of fraud possibility of a participant in a race from training information, including race state information about the race, by learning using race state information indicating the state of a past race extracted from measurement data on the past race; and an output unit 140 that outputs the degree of fraud possibility of the participant.

Description

[Technical Field]

[0001] The present disclosure relates to techniques for estimating fraud. [Background technology]

[0002] In publicly managed betting games, there are cases of match-fixing and other fraudulent activities. Situations that appear to be fraudulent are judged manually based on data such as odds and winning percentages.

[0003] Patent Document 1 describes that in a system for evaluating betting prediction data, it is possible to identify irregularities such as match-fixing based on the statistical significance between betting predictions and specific competition results. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2011-043916 Summary of the Invention [Problem to be solved by the invention]

[0005] With the technology of Patent Document 1, if the race results differ from the predictions made in advance, cheating may be pointed out regardless of whether or not actual cheating occurred. Furthermore, the technology of Patent Document 1 cannot be said to eliminate manual judgment.

[0006] One of the purposes of the present disclosure is to provide a fraud estimation device and the like that can reduce manual work. [Means for solving the problem]

[0007] A cheating estimation device according to one aspect of the present disclosure comprises an estimation means for estimating the degree of possibility of cheating by a participant in a target race from estimation information including race state information extracted from measurement data in the target race, using an estimation model generated by learning using race state information representing the state of the past race extracted from measurement data in the past race to estimate the degree of possibility of cheating by a participant in the race from learning information including race state information of the race, and an output means for outputting the degree of possibility of cheating by the participant.

[0008] A cheating estimation method according to one aspect of the present disclosure estimates the degree of possibility of cheating by a participant in a target race from estimation information including race state information extracted from measurement data in the target race using an estimation model generated by learning using race state information representing the state of the past race extracted from measurement data in the past race, to estimate the degree of possibility of cheating by the participant in the race from learning information including race state information of the race, and outputs the degree of possibility of cheating by the participant.

[0009] A storage medium according to one embodiment of the present disclosure stores a program that causes a computer to execute an estimation process that estimates the degree of possibility of cheating by a participant in a target race from estimation information including race state information extracted from measurement data in the target race, using an estimation model that is generated by learning using learning information including race state information that represents the state of the past race extracted from measurement data in the past race, to estimate the degree of possibility of cheating by a participant in the race from the race state information of the race, and an output process that outputs the degree of possibility of cheating by the participant.

[0010] A learning device according to one aspect of the present disclosure includes an extraction means for extracting race state information representing the state of a past race from measurement data of the past race, and a model generation means for generating an estimation model by learning using learning information including the race state information, to estimate the degree of possibility of cheating by participants in the race from estimation information including the race state information of the race.

[0011] A learning method according to one aspect of the present disclosure extracts race state information representing the state of a past race from measurement data of the past race, and generates an estimation model by learning using learning information including the race state information to estimate the degree of possibility of cheating by participants in the race from estimation information including the race state information of the race.

[0012] A storage medium according to one aspect of the present disclosure stores a program that causes a computer to execute an extraction process that extracts race state information representing the state of a past race from measurement data from the past race, and a model generation process that generates an estimation model from estimation information including the race state information of the race by learning using learning information including the race state information to estimate the degree of possibility of cheating by participants in the race.

[0013] One aspect of the present disclosure is also realized by a program stored on the above-mentioned storage medium. [Effects of the Invention]

[0014] The present disclosure has the effect of reducing manual work. [Brief explanation of the drawings]

[0015] [Figure 1] FIG. 1 is a block diagram illustrating an example of the configuration of a fraud estimation device according to a first embodiment of the present disclosure. [Figure 2] FIG. 2 is a flowchart illustrating an example of the operation of the fraud estimation device according to the first embodiment of the present disclosure. [Figure 3] FIG. 3 is a block diagram illustrating an example of the configuration of a learning device according to the second embodiment of the present disclosure. [Figure 4] FIG. 4 is a flowchart illustrating an example of the operation of the learning device according to the third embodiment of the present disclosure. [Figure 5] FIG. 5 is a block diagram illustrating an example of the configuration of a fraud estimation system according to the third embodiment of the present disclosure. [Figure 6] FIG. 6 is a block diagram illustrating an example of the configuration of a fraud estimation device according to the third embodiment of the present disclosure. [Figure 7] FIG. 7 is a block diagram illustrating an example of the configuration of a learning device according to the third embodiment of the present disclosure. [Figure 8] FIG. 8 is a flowchart illustrating an example of the operation of the learning device according to the third embodiment of the present disclosure. [Figure 9] FIG. 9 is a flowchart illustrating an example of the operation of the fraud estimation device according to the third embodiment of the present disclosure. [Figure 10] FIG. 10 is a diagram illustrating an example of a hardware configuration of a computer capable of realizing the fraud estimation device and the learning device according to an embodiment of the present disclosure. DETAILED DESCRIPTION OF THE INVENTION

[0016] Hereinafter, embodiments of the present disclosure will be described.

[0017] First Embodiment First, a first embodiment of the present disclosure will be described in detail with reference to the drawings.

[0018] <Configuration> FIG. 1 is a block diagram illustrating an example of the configuration of a cheating estimation device according to a first embodiment of the present disclosure. In the example illustrated in FIG. 1, the cheating estimation device 10 of this embodiment includes an estimation unit 130 and an output unit 140. The estimation unit 130 uses an estimation model to estimate the degree of likelihood of cheating by a participant in a target race from estimation information including race state information extracted from measurement data of the target race. The estimation model is generated by learning using learning information including race state information representing the state of past races extracted from measurement data of the past races, so as to estimate the degree of likelihood of cheating by a participant in the race from the race state information of the race. The output unit 140 outputs the degree of likelihood of cheating by the participant.

[0019] The race may be, for example, a boat race. The race may also be, for example, another publicly managed sport such as horse racing, bicycle racing, or auto racing. The target race is a race in which the degree of likelihood of participants cheating is estimated. The past races are one or more races of the same type as the target race that have been held in the past (for example, multiple races of the same type as the target race that have been held in the past).

[0020] The measurement data may be, for example, images such as video captured by at least one of an imaging device fixed to the racetrack and an imaging device mounted on a drone. The race state information may be, for example, a combination of the position and speed of a participant during a race. The measurement data may be a signal representing the state of an object (e.g., an accelerator or brake) operated by a participant in the race, measured by a sensor attached to the object. The race state information may be, for example, a transition of an operation by a participant in the race, estimated from the state of the object under investigation. Other examples of measurement data and race state information will be described in detail later.

[0021] The estimation model is, for example, an estimator configured to receive estimation information as input, estimate the degree of cheating by a contestant from the received estimation information, and output the estimated degree of cheating by the contestant. The estimator may be implemented, for example, as a processor that executes a program that implements the functions of the estimator. The estimator may be implemented, for example, as a dedicated circuit that implements the functions of the estimator. The estimator may be implemented, for example, as a combination of a processor that executes a program and a dedicated circuit that implements the functions of the estimator.

[0022] The estimation model is generated by the above-described learning. As a method for learning the estimation model, various existing learning methods including, for example, heterogeneous mixture learning can be used.

[0023] The degree of likelihood of a contestant cheating may be represented by either a value representing cheating or a value representing non-cheating. The degree of likelihood of a contestant cheating may be represented by a value greater than or equal to a lower limit and less than or equal to an upper limit. The degree of likelihood of a contestant cheating may be represented by any one of three or more values ​​including the upper limit and the lower limit.

[0024] <Operation> 2 is a flowchart illustrating an example of the operation of the cheating estimation device 10 according to the first embodiment of the present disclosure. In the example shown in FIG. 2, the estimation unit 130 uses an estimation model to estimate the degree of likelihood of cheating by a participant in a target race from estimation information including race state information extracted from estimation data for the target race (step S11). Then, the output unit 140 outputs the degree of cheating by the participant estimated by the estimation unit 130 (step S12).

[0025] <Effects> This embodiment has the advantage of reducing manual work, because the estimation unit 130 uses an estimation model to estimate the degree of likelihood of cheating by participants in a target race from the information for estimation.

[0026] To identify fraud, for example, information on races where the likelihood of fraud is higher than a predetermined standard can be checked, and information on races where the likelihood of fraud is lower than the predetermined standard can be omitted, thereby reducing manual work.

[0027] <Information for learning and estimation> The learning information includes the same types of information as the estimation information. The learning information may further include, for each past race, information identifying participants who committed cheating in past races. Cheating by participants in past races may be manually determined. The learning information may include information that indicates cheating in the race condition information and pre-race condition information of past races.

[0028] The learning information and the estimation information may each include pre-race state information extracted from measurement data of the contestant before the race (pre-race measurement data), in addition to race state information extracted from measurement data of the contestant before the race (hereinafter also referred to as race measurement data). The race state information may be the progression of a combination of one or more state values ​​each representing the state of the contestant measured during the race. The race state information is generated so that one or more state values ​​included in one combination represent the state of the contestant at the same time. The pre-race state information may be the progression of a combination of one or more state values ​​each representing the state of the contestant measured before the race. The pre-race state information is generated so that one or more state values ​​included in one combination represent the state of the contestant at the same time.

[0029] The learning information may include race information. The learning information may include, for example, a combination of odds information and race result information. The estimation information may include the estimation information. The estimation information may include odds information. The learning information and estimation information may each include a trend in the bias in the number of votes.

[0030] The training information and the estimation information may include status values ​​representing the attributes of the contestant. The status values ​​representing the attributes of the contestant may include, for example, information about the branch to which the contestant belongs and information indicating whether the stadium where the race is held is the contestant's home stadium. The status values ​​representing the attributes of the contestant may also include information representing the relationship between the contestant and other contestants. Information representing the relationship between the contestant and other contestants may include, for example, information representing a master-disciple relationship, information representing siblings, information representing friendships, information indicating whether the contestants attended the same training school, and the like. This information may be used, for example, to extract from the training information a pattern that is likely to occur in the transition of the status values ​​of the race status information when there is a specific relationship between the contestants and cheating is occurring. If the extracted pattern exists in the estimation information, the extracted pattern may be used to estimate the degree of possibility of cheating depending on the strength of the existing pattern.

[0031] <Race status information> As described above, the race status information is, for example, a transition of a combination of one or more status values ​​each representing the status of a participant measured during a race. The race measurement data is, as described above, images (e.g., multiple images or video) obtained by capturing the race with one or more imaging devices. The race status information is a transition of a combination of the position and speed of each participant extracted from the images captured by capturing the race. In this case, the position and speed are each the above-mentioned status values. If the competition is a boat race, the position of a participant may be the position of the boat on which the participant rides. The position of the boat is the position of a point appropriately defined on the boat. If the competition is another type of competition, the position of a participant may be the position of a point appropriately set on the device on which the participant rides. The device on which the participant rides in a competition is referred to as the racing device. Examples of the racing device include a boat, a motorcycle, a bicycle, etc. The position of a participant may be acquired, for example, by a position acquisition device that uses a distance acquisition system such as a Global Positioning System (GPS) installed on the racing device on which the participant rides. In this case, the measurement data may be, for example, an image and data representing the position output from a position acquisition device. The speed of the playing device may be a speed measured by a speedometer of the playing device.

[0032] The measurement data may include, for example, the acceleration of the racing device (i.e., the acceleration of the contestant) measured by an acceleration sensor mounted on the racing device as the above-mentioned state value. In this case, the race state information may include the transition of a combination of position and acceleration. The race state information may include the transition of a combination of position, speed, and acceleration. In this case, the state values ​​are position, speed, and acceleration.

[0033] The race state information may include, as the above-mentioned state value, posture information representing the posture of the contestants, recognized using existing image recognition technology from measurement data, which is an image of the race. The value of the posture information may be defined appropriately depending on the posture. The race state information may also include, as the above-mentioned state value, the output of a prime mover, such as a motor, of the racing device that the contestant rides during the race, measured by a sensor that measures the output of the prime mover (hereinafter referred to as prime mover output).

[0034] The measurement data may be a signal representing the state of an operation object (e.g., an accelerator) operated by a contestant, output by a sensor attached to the operation object and measuring the state of the operation object. The race state information may include a transition of a state value representing information about the operation object (e.g., a value representing the opening degree of the accelerator or a value representing the depression of the accelerator), which is generated from the signal representing the state of the operation object. In this case, the race state information may include, for example, a transition of a combination of a position and a value representing the depression of the accelerator.

[0035] The race state information may include, as a state value, at least one of a relative position and a relative speed of the two contestants for at least any combination of two contestants selected from the contestants.The race state information may include, as a state value, at least one of a relative position and a relative speed of the two contestants for each combination of two contestants selected from the contestants.

[0036] The combination of state values ​​included in the race state information is not limited to the above examples. The race state information may be a transition of a combination including at least one of the state values ​​listed above. Furthermore, the state values ​​are not limited to the above examples.

[0037] The measurement data obtained by measurements using a measuring device such as a sensor mounted on the sports device may be stored in a storage device provided in the sports device. The measurement data obtained by measurements using a measuring device such as a sensor mounted on the sports device may be collected by a communication device mounted on the sports device, for example, via wireless communication.

[0038] <Pre-race condition information> As described above, the pre-race condition information is, for example, a transition of a combination of one or more condition values ​​each representing the condition of a participant measured before the race.

[0039] The pre-race condition information may include the participant's biometric information measured before the race (e.g., heart rate measured by a heart rate monitor, blood pressure measured by a blood pressure monitor, etc.) The pre-race condition information may also include information extracted from the participant's biometric information measured before the race (e.g., the participant's eye movements extracted from a facial image of the participant).

[0040] The pre-race status information may include status values ​​extracted from images taken of events related to participants that are held before the race (for example, the start exhibition in a boat race, and the paddock in a horse race).

[0041] The pre-race status information may include a status value representing the behavior of the contestant before the race. The status value representing the behavior of the contestant before the race may be, for example, a value representing the behavior of the contestant extracted from an image of the contestant taken at a location where the contestant can perform a behavior before the race, for example, using an existing technology for estimating behavior. The value representing the behavior of the contestant may be a predetermined value for a predefined type of behavior. The status value representing the behavior of the contestant before the race may be, for example, a value representing the degree of suspiciousness of the contestant's behavior, extracted from an image of the contestant taken at a location where the contestant can perform a behavior before the race, for example, using an existing technology for estimating suspiciousness of behavior. The status value representing the behavior of the contestant before the race may be, for example, a value representing the emotional state of the contestant extracted from an image of the contestant taken at a location where the contestant can perform a behavior before the race, for example, using an existing technology for estimating emotions from behavior. The value representing the emotional state may be, for example, one of predetermined values ​​for each of a plurality of predetermined emotions.

[0042] <Estimation model> The estimation model may be generated to include a function for estimating the degree of likelihood of a contestant cheating related to a detected pattern, depending on the strength of the detected pattern, when a predetermined pattern is detected in the race state information. This predetermined pattern may be, for example, decelerating on a straight course, large fluctuations in ranking, excessive acceleration in a corner resulting in a wide turn, unnatural course selection (for example, taking a course that does not attack the inside when a leading contestant is widening in a corner), etc.

[0043] <Second embodiment> Next, a second embodiment of the present disclosure will be described in detail with reference to the drawings.

[0044] <Configuration> FIG. 3 is a block diagram illustrating an example of the configuration of a learning device according to a second embodiment of the present disclosure. In the example illustrated in FIG. 3, the learning device 20 includes an extraction unit 220 and a model generation unit 230. The extraction unit 220 extracts race state information representing the state of a past race from measurement data of the past race. The model generation unit 230 generates an estimation model by learning using learning information including the race state information to estimate the degree of likelihood of cheating by a participant in a race from estimation information including the race state information of the race. Specifically, the extraction unit 220 may extract the above-mentioned state values ​​from measurement data of a past race and generate race state information using the extracted state values.

[0045] <Operation> 4 is a flowchart illustrating an example of the operation of the learning device 20 according to the third embodiment of the present disclosure. In the example illustrated in FIG. 4, the extraction unit 220 extracts learning information including race state information from measurement data of past races (step S21). Next, the model generation unit 230 generates an estimation model by learning using the learning information so as to estimate the degree of likelihood of cheating by participants in a race from estimation information including race state information of the race (step S22).

[0046] <Effects> The present embodiment described above has the same effect as the first embodiment because the model generation unit 230 generates an estimation model by learning using learning information including race state information, so as to estimate the degree of possibility of cheating by participants in a race from estimation information including race state information of the race.

[0047] <Third embodiment> Next, a third embodiment of the present disclosure will be described in detail with reference to the drawings.

[0048] <Configuration> Fig. 5 is a block diagram illustrating an example of the configuration of a fraud estimation system according to a third embodiment of the present disclosure. In the example illustrated in Fig. 5, the fraud estimation system 1 includes a fraud estimation device 100, a learning device 200, a measurement device 300, a data storage device 400, and an output destination device 500. The fraud estimation device 100 is communicatively connected to each of the learning device 200, the measurement device 300, and the output destination device 500. The learning device 200 is communicatively connected to each of the fraud estimation device 100 and the data storage device 400. The data storage device 400 is communicatively connected to each of the measurement device 300 and the learning device 200.

[0049] In this embodiment, the race state information, pre-race state information, race information, and participant attributes of past races are the same as the race state information, pre-race state information, race information, and participant attributes of past races described above. Furthermore, the race state information, pre-race state information, race information, and participant attributes of the target race are also the same as the race state information, pre-race state information, race information, and participant attributes of the target race described above. The race information of the estimation information for the target race and the attributes of the participants for the target race are provided to the fraud estimation device 100 in advance.

[0050] <Measuring device 300> The measurement device 300 is a device that measures the race and the participants before the race. The measurement device 300 may include, for example, the above-mentioned imaging device that captures images of the race, the imaging device that captures images of the participants before the race, sensors installed on the racing equipment, and devices that receive the above-mentioned measurement data from the sensors. The sensors installed on the racing equipment may be any of the above-mentioned sensors, such as a position acquisition device that measures the position of the participants, an acceleration sensor, a speedometer, a sensor that measures an object to be operated such as an accelerator, etc.

[0051] The measurement device 300 transmits the measurement data obtained by the measurement device 300 (i.e., the measurement data of the race and the measurement data of the participants before the race) to the data storage device 400. The measurement device 300 also transmits the measurement data obtained by the measurement device 300 (i.e., the measurement data of the race and the measurement data of the participants before the race) to the fraud estimation device 100.

[0052] <Data storage device 400> The data storage device 400 receives measurement data of past races (i.e., measurement data of the race and measurement data of participants before the race) obtained by the measurement device 300 and stores the received measurement data. The data storage device 400 also stores race information of past races and attributes of participants in past races. The race information of past races and attributes of participants in past races are input into the data storage device 400 by, for example, a user of the fraud estimation system 1.

[0053] <Output destination device 500> The destination device 500 may be, for example, either an information processing device or a storage device, or may be another device.

[0054] <Fraud estimation device 100> 6 is a block diagram illustrating an example of the configuration of a fraud inference device 100 according to the third embodiment of the present disclosure. In the example illustrated in FIG. 6, the fraud inference device 100 includes a target data receiving unit 110, an extracting unit 120, an estimating unit 130, an output unit 140, a model accepting unit 150, and a model storage unit 160.

[0055] <Model Reception Department 150> The model receiving unit 150 receives information about the above-described estimation model from the learning device 200. The estimation model of this embodiment is the same as the estimation model of the first embodiment. In the description of this embodiment, the information about the estimation model is, for example, information necessary to operate a processor as the estimation model, including a program and parameters that realize the estimation model. The model receiving unit 150 stores the received information about the estimation model in the model storage unit 160.

[0056] <Model storage unit 160> The model storage unit 160 stores information about the estimation model. The information about the estimation model is read by the estimation unit 130, for example, before the cheating estimation device 100 estimates the degree of possibility of cheating by a participant in a target race.

[0057] <Target Data Receiving Unit 110> The target data receiving unit 110 receives measurement data of a target race (i.e., measurement data of the race and measurement data of participants before the race) from the measurement device 300. The target race is a race in which the degree of possibility of participants cheating is to be estimated. The target data receiving unit 110 sends the received measurement data of the target race to the extraction unit 120.

[0058] <Extraction part 120> The extraction unit 120 receives measurement data for the target race (i.e., measurement data for the target race and measurement data for participants prior to the target race) from the target data receiving unit 110. The extraction unit 120 extracts the above-mentioned state values ​​from the received measurement data for the target race using an existing method for extracting the above-mentioned state values. The extraction unit 120 generates race state information using the state values ​​extracted from the measurement data of the target race. The extraction unit 120 extracts the above-mentioned state values ​​from the received measurement data of participants prior to the target race using an existing method for extracting the above-mentioned state values. The extraction unit 120 generates pre-race state information using the state values ​​extracted from the measurement data of participants prior to the target race. The extraction unit 120 sends the race state information and the pre-race state information to the estimation unit 130.

[0059] <Estimation part 130> The estimation unit 130 receives race state information and pre-race state information from the extraction unit 120. In addition, the estimation unit 130 is provided with race information for estimation of the target race and attributes of the participants in the target race by, for example, the user of the cheating estimation device 100. In addition, when the cheating estimation device 100 starts operating, the estimation unit 130 reads out information about the estimation model from the model storage unit 160.

[0060] The estimation unit 130 estimates the degree of possibility of cheating by a participant in the target race, similar to the estimation unit 130 in the first embodiment. Specifically, the estimation unit 130 uses an estimation model to estimate the degree of possibility of cheating by a participant in the target race from estimation information including race state information and pre-race state information of the target race. The degree of possibility of cheating by a participant in the target race represents the degree of possibility that a participant in the target race has engaged in cheating during the target race.

[0061] <Output unit 140> The output unit 140 receives the degree of possibility of cheating by the participants in the target race from the estimation unit 130. The output unit 140 outputs the received degree of possibility of cheating by the participants in the target race.

[0062] <Learning device 200> 7 is a block diagram illustrating an example of the configuration of a learning device 200 according to the third embodiment of the present disclosure. In the example illustrated in FIG. 7, the learning device 200 includes a data acquisition unit 210, an extraction unit 220, a model generation unit 230, and a model output unit 240.

[0063] <Data Acquisition Unit 210> The data acquisition unit 210 acquires measurement data of past races (specifically, measurement data of past races and measurement data of contestants before past races), race information, and contestant attributes from the data storage device 400. That is, the data acquisition unit 210 reads out the measurement data, race information, and contestant attributes of past races from the data storage device 400. The data acquisition unit 210 sends the read measurement data, race information, and contestant attributes of past races to the extraction unit 220.

[0064] <Extraction part 220> The extraction unit 220 receives measurement data of past races (specifically, measurement data of past races and measurement data of contestants before the past races), race information, and contestant attributes from the data acquisition unit 210. The extraction unit 220 extracts the above-mentioned state values ​​from the measurement data of past races and generates race state information and pre-race state information for past races using the extracted state values. Specifically, the extraction unit 220 extracts state values ​​from the measurement data of past races and generates race state information using the state values ​​extracted from the measurement data of past races. The extraction unit 220 extracts state values ​​from the measurement data of contestants before the past races and generates pre-race state information using the state values ​​extracted from the measurement data of contestants before the past races. The method by which the extraction unit 220 extracts state values ​​is the same as the method by which the extraction unit 120 of the fraud estimation device 100 extracts state values.

[0065] The extraction unit 220 sends the generated race state information and pre-race state information of past races, and the received race information and participant attributes of past races to the model generation unit 230.

[0066] <Model Generation Unit 230> The model generation unit 230 receives race state information, pre-race state information, race information, and participant attributes of past races from the extraction unit 220. The model generation unit 230 generates an estimation model by learning using the received race state information, pre-race state information, race information, and participant attributes of past races (i.e., learning information).

[0067] The model generation unit 230 sends information about the generated estimation model to the model output unit 240.

[0068] <Model Output Unit 240> The model output unit 240 receives information about the estimated model from the model generation unit 230. The model output unit 240 transmits the information about the estimated model to the fraud inference device 100.

[0069] <Operation> Next, the operations of the fraud estimation device 100 and the learning device 200 according to the third embodiment of the present disclosure will be described in detail with reference to the drawings.

[0070] FIG. 8 is a flowchart illustrating an example of the operation of the learning device 200 according to the third embodiment of the present disclosure. In the example illustrated in FIG. 8, the data acquisition unit 210 of the learning device 200 acquires measurement data of a past race (step S101). The data acquisition unit 210 acquires measurement data of a participant before the past race (step S102). The data acquisition unit 210 may perform the operation of step S101 and the operation of step S102 simultaneously. The data acquisition unit 210 may perform the operation of step S101 and the operation of step S102 in parallel. The data acquisition unit 210 may perform the operation of step S101 and the operation of step S102 in reverse order. The data acquisition unit 210 may perform a combination of the operation of step S101 and the operation of step S102 for each past race. The data acquisition unit 210 may acquire race information and participant attributes of the past race in the operation of step S101 or the operation of step S102.

[0071] Next, the extraction unit 220 extracts race state information from the measurement data of the past race (step S103). Specifically, as described above, the extraction unit 220 extracts state values ​​from the measurement data of the past race, and generates race state information using the state values ​​extracted from the measurement data of the past race.

[0072] The extraction unit 220 further extracts pre-race condition information from the measurement data of the contestant before the past race (step S104). Specifically, the extraction unit 220 extracts condition values ​​from the measurement data of the contestant before the past race, and generates pre-race condition information using the condition values ​​extracted from the measurement data of the contestant before the past race. The data acquisition unit 210 may perform the operations of steps S103 and S104 simultaneously. The data acquisition unit 210 may perform the operations of steps S103 and S104 in parallel. The data acquisition unit 210 may perform the operations of steps S103 and S104 in the reverse order. The data acquisition unit 210 may perform a combination of the operations of steps S103 and S104 for each past race.

[0073] The model generation unit 230 generates an estimation model by learning using the race state information and pre-race information (step S105). The model generation unit 230 may generate an estimation model by learning using race information and attributes of participants in addition to the race state information and pre-race information.

[0074] The model output unit 240 outputs the estimation model (Step S106). Specifically, the model output unit 240 outputs information on the generated estimation model to the fraud inference device 100.

[0075] 9 is a flowchart illustrating an example of the operation of the fraud estimation device 100 according to the third embodiment of the present disclosure. Before starting the operation of FIG. 9, the model reception unit 150 receives information about the estimation model from the learning device 200 and stores the received information about the estimation model in the model storage unit 160. Then, the estimation unit 130 reads out the information about the estimation model from the model storage unit 160.

[0076] 9, the target data receiving unit 110 receives measurement data of participants before the target race (step S111). The target data receiving unit 110 further receives measurement data of the target race (step S112).

[0077] The extraction unit 120 extracts pre-race condition information for the target race from the measurement data of the participants before the target race (step S113). Specifically, the extraction unit 120 extracts condition values ​​from the measurement data of the participants before the target race, and generates pre-race condition information for the target race using the condition values ​​extracted from the measurement data of the participants before the target race.

[0078] The extraction unit 120 further extracts race state information for the target race from the measurement data for the target race (step S114). Specifically, the extraction unit 120 extracts state values ​​from the measurement data for the target race, and generates race state information for the target race using the state values ​​extracted from the measurement data for the target race.

[0079] Next, the estimation unit 130 uses the estimation model to estimate the degree of possibility of cheating by the participants in the target race from the state information of the target race and the pre-race state information (step S115). In step S115, the estimation unit 130 may estimate the degree of possibility of cheating by the participants in the target race from the state information of the target race, the pre-race state information, the race information, and the attributes of the participants.

[0080] Next, the output unit 140 outputs the estimated degree of possibility of cheating by the participants in the target race (step S116).

[0081] <Effects> The present embodiment described above has the same effects as the first embodiment. The reason for this is the same as the reason for the effects of the first embodiment. In addition, the present embodiment also has the same effects as the second embodiment for the same reasons. <First Modification of the Second Embodiment> This modification is the same as the second embodiment except for the differences described below.

[0082] In this modification, the model generation unit 230 generates an estimation model using a learning method, such as heterogeneous mixture learning, that can generate a model that can further estimate the factors that contribute to the estimation result and the magnitude of the contribution of those factors.

[0083] The estimation unit 130 uses such an estimation model to further estimate factors that contribute to the result of the estimation (i.e., the degree of possibility of cheating by a participant in the target race) and the magnitude of the contribution of the factors.

[0084] The output unit 140 outputs the degree of possibility of cheating by the participant in the target race, as well as the factors that contribute to the degree of possibility of cheating by the participant in the target race and the magnitude of the contribution of each factor.

[0085] <Second Modification of the Second Embodiment> This modification is the same as the second embodiment except for the differences described below.

[0086] The data storage device 400 may further store race condition information and pre-race condition information of past races extracted from the measurement data obtained by the measurement device 300.

[0087] The data acquisition unit 210 of the learning device 200 reads out the measurement data, race information, and participant attributes for past races for which race state information and pre-race state information have not been accumulated. For past races for which race state information and pre-race state information have been accumulated, the data acquisition unit 210 reads out the accumulated race state information, pre-race state information, race information, and participant attributes.

[0088] The extraction unit 220 generates race state information and pre-race state information from the read measurement data, and stores the generated race state information and pre-race state information in the data storage device 400.

[0089] <Another Modification of the Second Embodiment> The fraud estimation device 100 and the learning device 200 may be implemented as the same device.

[0090] Furthermore, the target data receiving unit 110 of the fraud estimation device 100 may receive measurement data of participants before the target race before the target race is held. The extraction unit 120 may extract pre-race state information from the measurement data of participants before the target race before the target race is held. The estimation unit 130 may estimate the degree of likelihood that participants in the target race will commit fraud from the pre-race state information of the target race using an estimation model that estimates the degree of likelihood that participants will commit fraud from the pre-race state information before the target race is held. The output unit 140 may output the degree of likelihood that participants in the target race will commit fraud before the target race is held. In this case, the output destination device 500 may be an information processing device or display device that can be viewed by the organizer or related parties of the target race. In this case, the output destination device 500 may be an information processing device or display device that displays the received information on a screen to people within a range that can view the screen of the output destination device 500. In this case, the output destination device 500 may be an information processing device that distributes information about the target race. The information processing device may distribute information indicating the degree of likelihood that participants in the target race will commit fraud to terminals of registered users who are registered as recipients of the information.

[0091] <Other embodiments> The fraud estimation device and learning device according to the above-described embodiments of the present disclosure can be realized by a computer including a memory into which a program read from a storage medium is loaded and a processor that executes the program. The fraud estimation device and learning device according to the above-described embodiments can also be realized by dedicated hardware. The fraud estimation device and learning device according to the above-described embodiments can also be realized by a combination of the above-described computer and dedicated hardware.

[0092] FIG. 10 is a diagram illustrating an example of the hardware configuration of a computer 1000 that can realize the fraud estimation device and the learning device according to the above-described embodiment. In the example illustrated in FIG. 10, the computer 1000 includes a processor 1001, a memory 1002, a storage device 1003, and an I / O (Input / Output) interface 1004. The computer 1000 can also access a storage medium 1005. The memory 1002 and the storage device 1003 are, for example, storage devices such as RAM (Random Access Memory) and a hard disk. The storage medium 1005 is, for example, a storage device such as RAM or a hard disk, a ROM (Read Only Memory), or a portable storage medium. The storage device 1003 may also be the storage medium 1005. The processor 1001 can read and write data and programs from and to the memory 1002 and the storage device 1003. The processor 1001 can access other devices via the I / O interface 1004. The processor 1001 can also access the storage medium 1005. The storage medium 1005 stores a program that causes the computer 1000 to operate as the fraud estimation device according to the above-described embodiment, or a program that causes the computer 1000 to operate as the learning device according to the above-described embodiment.

[0093] The processor 1001 loads a program stored in the storage medium 1005, which causes the computer 1000 to operate as the fraud estimation device according to the above-described embodiment, into the memory 1002. Then, the processor 1001 executes the program loaded into the memory 1002, causing the computer 1000 to operate as the fraud estimation device according to the above-described embodiment.

[0094] The processor 1001 loads a program stored in the storage medium 1005, which causes the computer 1000 to operate as the learning device according to the above-described embodiment, into the memory 1002. The processor 1001 then executes the program loaded into the memory 1002, causing the computer 1000 to operate as the learning device according to the above-described embodiment.

[0095] The target data receiving unit 110, the extraction unit 120, the estimation unit 130, the output unit 140, and the model receiving unit 150 can be realized, for example, by a processor 1001 that executes a program loaded into a memory 1002. The data acquisition unit 210, the extraction unit 220, the model generation unit 230, and the model output unit 240 can be realized, for example, by a processor 1001 that executes a program loaded into a memory 1002. The model storage unit 160 can be realized by a memory 1002 or a storage device 1003 such as a hard disk drive included in the computer 1000. Some or all of the target data receiving unit 110, the extraction unit 120, the estimation unit 130, the output unit 140, the model reception unit 150, and the model storage unit 160 can be realized by dedicated circuits that realize the functions of each unit. Some or all of the data acquisition unit 210, the extraction unit 220, the model generation unit 230, and the model output unit 240 can be realized by dedicated circuits that realize the functions of each unit.

[0096] Furthermore, some or all of the above-described embodiments can be described as, but are not limited to, the following supplementary notes.

[0097] (Appendix 1) an estimation means for estimating the degree of possibility of cheating by a participant in a target race from estimation information including race state information extracted from measurement data in the target race, using an estimation model generated by learning using learning information including race state information that represents the state of the past race extracted from measurement data in the past race, so as to estimate the degree of possibility of cheating by a participant in the race from race state information; an output means for outputting the degree of possibility of cheating by the contestant; A fraud estimation device comprising:

[0098] (Appendix 2) The race state information includes a transition of a combination of a position and a speed of each participant in the race, extracted from an image obtained as the measurement data by capturing an image of the race. 2. The fraud estimation device according to claim 1.

[0099] (Appendix 3) The race state information includes a transition of a combination of relative positions and relative speeds between participants in the race, which is extracted from images obtained as the measurement data by capturing images of the race. 3. The fraud estimation device according to claim 1 or 2.

[0100] (Appendix 4) The race state information includes a transition of an estimated operation amount extracted from measurement data of an operation object operated by the contestant. 4. The fraud estimation device according to any one of appendices 1 to 3.

[0101] (Appendix 5) the estimation model is generated by the learning using the learning information further including pre-race state information of past races so as to estimate the degree of cheating for each participant from pre-race state information extracted from the measurement data of the participants before the race and the race state information; The estimation means estimates the degree of possibility of the cheating using the estimation information including the pre-race condition information of the contestant in the target race before the target race. 5. The fraud estimation device according to any one of appendices 1 to 4.

[0102] (Appendix 6) The pre-race condition information includes biometric information extracted from the biometric data of the participant obtained as the measurement data. 6. The fraud estimation device according to claim 5.

[0103] (Appendix 7) The pre-race state information includes a progression of estimated actions of the contestant extracted from images of the contestant taken before the race. 7. The fraud estimation device according to claim 5 or 6.

[0104] (Appendix 8) The pre-race state information includes an estimated state transition of the contestant in the event extracted from images of the contestant in the event before the race. 8. The fraud estimation device according to any one of appendices 5 to 7.

[0105] (Appendix 9) the estimation model is generated by the learning so as to estimate factors that contribute to the degree of likelihood and the magnitude of contribution of the factors to the degree of likelihood; The estimation means uses the estimation model to estimate factors that contribute to the degree of the possibility of cheating by the participant in the target race and the magnitude of the contribution of the factors to the degree of the possibility; The output means further outputs the factors and the magnitude of the contribution of the factors. 9. A fraud estimation device according to any one of appendices 1 to 8.

[0106] (Appendix 10) a learning device including a learning means for performing the learning of the estimation model; A fraud estimation device according to any one of Supplementary Notes 1 to 9; A fraud estimation system including:

[0107] (Appendix 11) extraction means for extracting race state information representing the state of a past race from measurement data of the past race; a model generation means for generating an estimation model by learning using learning information including the race state information, so as to estimate the degree of possibility of cheating by a participant in the race from estimation information including the race state information of the race; A learning device comprising:

[0108] (Appendix 12) The race state information includes a transition of a combination of a position and a speed of each participant in the race, extracted from an image obtained as the measurement data by capturing an image of the race. 12. The learning device of claim 11.

[0109] (Appendix 13) The race state information includes a transition of a combination of relative positions and relative speeds between participants in the race, which is extracted from images obtained as the measurement data by capturing images of the race. 13. The learning device according to claim 11 or 12.

[0110] (Appendix 14) The race state information includes a transition of an estimated operation amount extracted from measurement data of an operation object operated by the contestant. 14. A learning device according to any one of appendices 11 to 13.

[0111] (Appendix 15) The model generation means generates the estimation model by learning using the learning information further including pre-race state information of past races so that the estimation model estimates the degree of cheating for each of the contestants from the estimation information further including pre-race state information representing the pre-race state of the contestants extracted from measurement data of the contestants before the race. A learning device according to any one of appendices 11 to 14.

[0112] (Appendix 16) The pre-race condition information includes biometric information extracted from the biometric data of the participant obtained as the measurement data. 16. The learning device of claim 15.

[0113] (Appendix 17) The pre-race state information includes a progression of estimated actions of the contestant extracted from images of the contestant taken before the race. 17. The learning device according to claim 15 or 16.

[0114] (Appendix 18) The pre-race state information includes an estimated state transition of the contestant in the event extracted from images of the contestant in the event before the race. 18. A learning device according to any one of appendices 15 to 17.

[0115] (Appendix 19) The model generation means generates the estimation model through the learning so that the estimation model estimates factors that contribute to the degree of likelihood and the magnitude of contribution of the factors to the degree of likelihood. A learning device according to any one of appendices 11 to 18.

[0116] (Appendix 20) using an estimation model generated by learning using learning information including race state information that represents the state of a past race extracted from measurement data of the past race, to estimate the degree of possibility of cheating by a participant in the race from race state information, the degree of possibility of cheating being committed by a participant in the target race is estimated from estimation information including race state information extracted from measurement data of the target race; outputting a degree of likelihood of cheating of the contestant; Fraud estimation method.

[0117] (Appendix 21) The race state information includes a transition of a combination of a position and a speed of each participant in the race, extracted from an image obtained as the measurement data by capturing an image of the race. The fraud estimation method described in Appendix 20.

[0118] (Appendix 22) The race state information includes a transition of a combination of relative positions and relative speeds between participants in the race, which is extracted from images obtained as the measurement data by capturing images of the race. 22. A fraud estimation method as set forth in appendix 20 or 21.

[0119] (Appendix 23) The race state information includes a transition of an estimated operation amount extracted from measurement data of an operation object operated by the contestant. 23. A fraud estimation method according to any one of appendices 20 to 22.

[0120] (Appendix 24) the estimation model is generated by the learning using the learning information further including pre-race state information of past races so as to estimate the degree of cheating for each participant from pre-race state information extracted from the measurement data of the participants before the race and the race state information; The degree of the possibility of the cheating is estimated using the information for estimation including the pre-race condition information of the participant of the target race before the target race. 24. A fraud estimation method according to any one of appendices 20 to 23.

[0121] (Appendix 25) The pre-race condition information includes biometric information extracted from the biometric data of the participant obtained as the measurement data. The fraud estimation method described in Appendix 24.

[0122] (Appendix 26) The pre-race state information includes a progression of estimated actions of the contestant extracted from images of the contestant taken before the race. 26. A fraud estimation method as set forth in Appendix 24 or 25.

[0123] (Appendix 27) The pre-race state information includes an estimated state transition of the contestant in the event extracted from images of the contestant in the event before the race. 27. A fraud estimation method according to any one of appendices 24 to 26.

[0124] (Appendix 28) the estimation model is generated by the learning so as to estimate factors that contribute to the degree of likelihood and the magnitude of contribution of the factors to the degree of likelihood; Using the estimation model, estimate factors that contribute to the degree of the possibility of cheating by the participant in the target race and the magnitude of the contribution of the factors to the degree of the possibility; sFurthermore, the factors and the magnitude of the contribution of the factors are output. A fraud estimation method according to any one of appendices 20 to 27.

[0125] (Appendix 29) extracting race state information representing the state of the past race from the measurement data of the past race; generating an estimation model by learning using learning information including the race state information, to estimate the degree of possibility of cheating by participants in the race from estimation information including the race state information of the race; How to learn.

[0126] (Appendix 30) The race state information includes a transition of a combination of a position and a speed of each participant in the race, extracted from an image obtained as the measurement data by capturing an image of the race. Study methods described in Appendix 29.

[0127] (Appendix 31) The race state information includes a transition of a combination of relative positions and relative speeds between participants in the race, which is extracted from images obtained as the measurement data by capturing images of the race. 2. A learning method as set forth in Appendix 29 or 30.

[0128] (Appendix 32) The race state information includes a transition of an estimated operation amount extracted from measurement data of an operation object operated by the contestant. 32. A learning method according to any one of appendices 29 to 31.

[0129] (Appendix 33) The estimation model is generated by learning using the learning information further including pre-race state information of past races so that the estimation model estimates the degree of cheating for each of the contestants from the estimation information further including pre-race state information representing the pre-race state of the contestants extracted from the measurement data of the contestants before the race. 33. A learning method according to any one of appendices 29 to 32.

[0130] (Appendix 34) The pre-race condition information includes biometric information extracted from the biometric data of the participant obtained as the measurement data. The study method described in Appendix 33.

[0131] (Appendix 35) The pre-race state information includes a progression of estimated actions of the contestant extracted from images of the contestant taken before the race. A learning method as described in Appendix 33 or 34.

[0132] (Appendix 36) The pre-race state information includes an estimated state transition of the contestant in the event extracted from images of the contestant in the event before the race. 36. A learning method according to any one of appendices 33 to 35.

[0133] (Appendix 37) generating the estimation model by the learning so that the estimation model estimates factors that contribute to the degree of likelihood and the magnitude of contribution of the factors to the degree of likelihood; 37. A learning method according to any one of appendices 29 to 36.

[0134] (Appendix 38) an estimation process for estimating the degree of possibility of cheating by a participant in a target race from estimation information including race state information extracted from measurement data in the target race, using an estimation model generated by learning using learning information including race state information that represents the state of the past race extracted from measurement data in the past race, so as to estimate the degree of possibility of cheating by a participant in the race from race state information; an output process for outputting the degree of possibility of cheating by the contestant; A storage medium that stores a program that causes a computer to execute the above.

[0135] (Appendix 39) The race state information includes a transition of a combination of a position and a speed of each participant in the race, extracted from an image obtained as the measurement data by capturing an image of the race. 39. The storage medium of claim 38.

[0136] (Appendix 40) The race state information includes a transition of a combination of relative positions and relative speeds between participants in the race, which is extracted from images obtained as the measurement data by capturing images of the race. 39. A storage medium according to claim 38 or 39.

[0137] (Appendix 41) The race state information includes a transition of an estimated operation amount extracted from measurement data of an operation object operated by the contestant. 41. A storage medium according to any one of appendices 38 to 40.

[0138] (Appendix 42) the estimation model is generated by the learning using the learning information further including pre-race state information of past races so as to estimate the degree of cheating for each participant from pre-race state information extracted from the measurement data of the participants before the race and the race state information; The estimation process estimates the degree of possibility of the cheating using the estimation information including the pre-race condition information of the contestant in the target race before the target race. 42. A storage medium according to any one of appendices 38 to 41.

[0139] (Appendix 43) The pre-race condition information includes biometric information extracted from the biometric data of the participant obtained as the measurement data. 43. The storage medium of claim 42.

[0140] (Appendix 44) The pre-race state information includes a progression of estimated actions of the contestant extracted from images of the contestant taken before the race. 44. A storage medium according to claim 42 or 43.

[0141] (Appendix 45) The pre-race state information includes an estimated state transition of the contestant in the event extracted from images of the contestant in the event before the race. 45. A storage medium according to any one of appendices 42 to 44.

[0142] (Appendix 46) the estimation model is generated by the learning so as to estimate factors that contribute to the degree of likelihood and the magnitude of contribution of the factors to the degree of likelihood; The estimation process uses the estimation model to estimate factors that contribute to the degree of the possibility of cheating by the participant in the target race and the magnitude of the contribution of the factors to the degree of the possibility; The output process further outputs the factors and the magnitude of the contribution of the factors. 46. ​​A storage medium according to any one of appendices 38 to 45.

[0143] (Appendix 47) an extraction process for extracting race state information representing the state of a past race from measurement data of the past race; a model generation process for generating an estimation model by learning using learning information including the race state information, so as to estimate the degree of possibility of cheating by participants in the race from estimation information including the race state information of the race; A storage medium that allows a computer to execute the above.

[0144] (Appendix 48) The race state information includes a transition of a combination of a position and a speed of each participant in the race, extracted from an image obtained as the measurement data by capturing an image of the race. 48. The storage medium of claim 47.

[0145] (Appendix 49) The race state information includes a transition of a combination of relative positions and relative speeds between participants in the race, which is extracted from images obtained as the measurement data by capturing images of the race. 49. A storage medium according to claim 47 or 48.

[0146] (Appendix 50) The race state information includes a transition of an estimated operation amount extracted from measurement data of an operation object operated by the contestant. 50. A storage medium according to any one of appendices 47 to 49.

[0147] (Appendix 51) The model generation process generates the estimation model by learning using the learning information further including pre-race state information of past races so that the estimation model estimates the degree of cheating for each of the contestants from the estimation information further including pre-race state information representing the pre-race state of the contestants extracted from measurement data of the contestants before the race. 51. A storage medium according to any one of appendices 47 to 50.

[0148] (Appendix 52) The pre-race condition information includes biometric information extracted from the biometric data of the participant obtained as the measurement data. 52. The storage medium of claim 51.

[0149] (Appendix 53) The pre-race state information includes a progression of estimated actions of the contestant extracted from images of the contestant taken before the race. 53. A storage medium according to claim 51 or 52.

[0150] (Appendix 54) The pre-race state information includes an estimated state transition of the contestant in the event extracted from images of the contestant in the event before the race. 54. A storage medium according to any one of appendices 51 to 53.

[0151] (Appendix 55) The model generation process generates the estimation model through the learning so that the estimation model estimates factors that contribute to the degree of likelihood and the magnitude of contribution of the factors to the degree of likelihood. 55. A storage medium according to any one of appendices 51 to 54.

[0152] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above embodiments. Various modifications that can be understood by those skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure. [Explanation of symbols]

[0153] 1. Fraud Prediction System 10. Fraud estimation device 20 Learning Device 100 Fraud Estimation Device 110 Target data receiving unit 120 Extraction part 130 Estimation part 140 Output section 150 Model Reception 160 Model Memory Unit 200 Learning Device 210 Data Acquisition Unit 220 Extraction part 230 Model Generation Unit 240 Model Output Section 300 Measuring Equipment 400 Data storage device 500 Output Device 1000 computers 1001 processor 1002 memory 1003 Storage device 1004 I / O interface 1005 Storage medium

Claims

1. an estimation means for estimating the degree of possibility of cheating by a participant in a target race from estimation information including race state information extracted from measurement data in the target race, using an estimation model generated by learning using learning information including race state information that represents the state of the past race extracted from measurement data in the past race, so as to estimate the degree of possibility of cheating by a participant in the race from race state information; an output means for outputting the degree of possibility of cheating by the contestant; A fraud estimation device comprising:

2. The race state information includes a transition of a combination of a position and a speed of each participant in the race, which is extracted from an image obtained as the measurement data by capturing an image of the race. The fraud estimation device according to claim 1 .

3. The race state information includes a transition of a combination of relative positions and relative speeds between participants in the race, which is extracted from images obtained as the measurement data by capturing images of the race. The fraud estimation device according to claim 1 or 2.

4. The race state information includes a transition of an estimated operation amount extracted from measurement data of an operation object operated by the contestant. The fraud estimation device according to any one of claims 1 to 3.

5. the estimation model is generated by the learning using the learning information further including pre-race state information of past races so as to estimate the degree of cheating for each participant from pre-race state information extracted from the measurement data of the participants before the race and the race state information; The estimation means estimates the degree of possibility of the cheating using the estimation information including the pre-race condition information of the contestant in the target race before the target race. The fraud estimation device according to any one of claims 1 to 4.

6. The pre-race condition information includes biometric information extracted from biometric data of the contestant obtained as the measurement data. The fraud estimation device according to claim 5 .

7. The pre-race state information includes a progression of estimated actions of the contestant extracted from images of the contestant taken before the race. The fraud estimation device according to claim 5 or 6.

8. The pre-race state information includes an estimated state transition of the contestant in the event extracted from images of the contestant in the event before the race. The fraud estimation device according to any one of claims 5 to 7.

9. A fraud estimation device, using an estimation model generated by learning using learning information including race state information that represents the state of a past race extracted from measurement data of the past race, to estimate the degree of possibility of cheating by a participant in the race from race state information, the degree of possibility of cheating being committed by a participant in the target race is estimated from estimation information including race state information extracted from measurement data of the target race; outputting a degree of likelihood of cheating of the contestant; Fraud estimation method.

10. an estimation process for estimating the degree of possibility of cheating by a participant in a target race from estimation information including race state information extracted from measurement data in the target race, using an estimation model generated by learning using learning information including race state information that represents the state of the past race extracted from measurement data in the past race, so as to estimate the degree of possibility of cheating by a participant in the race from race state information; an output process for outputting the degree of possibility of cheating by the contestant; A program that causes a computer to execute the following.

Images