BEHAVIOR PREDICTION DEVICE, BEHAVIOR PREDICTION METHOD, AND BEHAVIOR PREDICTION PROGRAM
A behavior prediction device is a behavior prediction device that predicts future behavior of a specific person, the behavior prediction device including: an information extraction unit that extracts start time probability distribution data indicating a probability distribution of a start time for each type of behavior, behavior transition probability data indicating a probability of transition from one behavior to the other behavior for each combination of the types of behaviors, and behavioral time data regarding a behavioral time of the each type of behavior on the basis of a series of past behaviors indicating a past behavior history of the specific person in chronological order; and a behavior prediction unit that predicts the future behavior of the specific person on the basis of the start time probability distribution data, the behavior transition probability data, the behavioral time data, and a current behavior of the specific person.
1 . A behavior prediction device that predicts future behavior of a specific person, the behavior prediction device comprising:
a memory; and
a processor configured to execute a process, the process including:
extracting start time probability distribution data indicating a probability distribution of a start time for each type of behavior, behavior transition probability data indicating a probability of transition from one behavior to the other behavior for each combination of the types of behaviors, and behavioral time data regarding a behavioral time of the each type of behavior on the basis of a series of past behaviors indicating a past behavior history of the specific person in chronological order; and
predicting the future behavior of the specific person on the basis of the start time probability distribution data, the behavior transition probability data, the behavioral time data, and a current behavior of the specific person.
2 . The behavior prediction device according to claim 1 , wherein
the behavioral time data includes behavioral time probability distribution data indicating a probability distribution of the behavioral time during which a behavior is performed for the each type of behavior and elapsed time probability distribution data indicating, for each type of next behavior, a probability distribution of elapsed time from start of one behavior to start of the other behavior for the each combination of the types of behaviors.
3 . The behavior prediction device according to claim 1 , further comprising:
estimating the current behavior of the specific person on the basis of a feature value obtained from a detection result of a sensor that detects an event that changes depending on a person's behavior.
4 . The behavior prediction device according to claim 1 , wherein:
extracting end time probability distribution data indicating a probability distribution of an end time for the each type of behavior on the basis of the series of past behaviors; and
predicting the future behavior of the specific person on the basis of the end time probability distribution data.
5 . A behavior prediction method of predicting future behavior of a specific person, the method comprising:
extracting start time probability distribution data indicating a probability distribution of a start time for each type of behavior, behavior transition probability data indicating a probability of transition from one behavior to the other behavior for each combination of the types of behaviors, and behavioral time data regarding a behavioral time of the each type of behavior on the basis of a series of past behaviors indicating a past behavior history of the specific person in chronological order; and
predicting the future behavior of the specific person on the basis of the start time probability distribution data, the behavior transition probability data, the behavioral time data, and a current behavior of the specific person.
6 . (canceled)
7 . The behavior prediction device according to claim 3 , wherein:
a feature extraction circuitry extracts features related to events that change due to the behavior of the specific person from the detection result of the sensor.
8 . The behavior prediction device according to claim 7 , wherein:
the feature extraction circuitry extracts vital sign, location, temperature, humidity, illuminance, sound volume, sleeping state, awake state, and electricity consumption of the specific person.
9 . The behavior prediction method according to claim 5 , wherein
the behavioral time data includes behavioral time probability distribution data indicating a probability distribution of the behavioral time during which a behavior is performed for the each type of behavior and elapsed time probability distribution data indicating, for each type of next behavior, a probability distribution of elapsed time from start of one behavior to start of the other behavior for the each combination of the types of behaviors.
10 . The behavior prediction method according to claim 5 , the method further comprising:
estimating the current behavior of the specific person on the basis of a feature value obtained from a detection result of a sensor that detects an event that changes depending on a person's behavior.
11 . The behavior prediction device according to claim 5 , the method further comprising:
extracting end time probability distribution data indicating a probability distribution of an end time for the each type of behavior on the basis of the series of past behaviors; and
predicting the future behavior of the specific person on the basis of the end time probability distribution data.
12 . The behavior prediction method according to claim 10 , wherein:
a feature extraction circuitry extracts features related to events that change due to the behavior of the specific person from the detection result of the sensor.
13 . The behavior prediction method according to claim 12 , wherein:
the feature extraction circuitry extracts vital sign, location, temperature, humidity, illuminance, sound volume, sleeping state, awake state, and electricity consumption of the specific person.
14 . A computer-readable non-transitory recording medium storing computer-executable program instructions that when executed by a processor cause a computer to execute a behavior prediction method comprising:
extracting start time probability distribution data indicating a probability distribution of a start time for each type of behavior, behavior transition probability data indicating a probability of transition from one behavior to the other behavior for each combination of the types of behaviors, and behavioral time data regarding a behavioral time of the each type of behavior on the basis of a series of past behaviors indicating a past behavior history of the specific person in chronological order; and
predicting the future behavior of the specific person on the basis of the start time probability distribution data, the behavior transition probability data, the behavioral time data, and a current behavior of the specific person.
15 . The computer-readable non-transitory recording medium according to claim 14 wherein the behavior prediction method further comprises:
behavioral time probability distribution data indicating a probability distribution of the behavioral time during which a behavior is performed for the each type of behavior and elapsed time probability distribution data indicating, for each type of next behavior, a probability distribution of elapsed time from start of one behavior to start of the other behavior for the each combination of the types of behaviors.
16 . The computer-readable non-transitory recording medium according to claim 14 wherein the behavior prediction method further comprises:
estimating the current behavior of the specific person on the basis of a feature value obtained from a detection result of a sensor that detects an event that changes depending on a person's behavior.
17 . The computer-readable non-transitory recording medium according to claim 14 wherein the behavior prediction method further comprises:
extracting end time probability distribution data indicating a probability distribution of an end time for the each type of behavior on the basis of the series of past behaviors; and
predicting the future behavior of the specific person on the basis of the end time probability distribution data.
18 . The computer-readable non-transitory recording medium according to claim 17 wherein the behavior prediction method further comprises:
extracting features related to events that change due to the behavior of the specific person from the detection result of the sensor.
19 . The computer-readable non-transitory recording medium according to claim 18 wherein the behavior prediction method further comprises:
extracting vital sign, location, temperature, humidity, illuminance, sound volume, sleeping state, awake state, and electricity consumption of the specific person.
20 . The behavior prediction device according to claim 1 , wherein:
a behavior estimation model is obtained by training a learning model.
21 . The behavior prediction device according to claim 20 , wherein:
the behavior estimation model is further trained by inputting a training data, vectorizing extracted features, and outputting an estimation result.