TERMINAL DEVICE, TERMINAL CONTROL METHOD, PROGRAM AND INFORMATION PROCESSING SYSTEM

• US 20130262354A1
• Filed: 02/26/2013
• Published: 10/03/2013
• Est. Priority Date: 03/30/2012
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Ide, Naoki, Usui, Takashi, Tsuda, Shinichiro
• Granted Patent: US 9,547,827 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06N 20/00 Machine learning

Abstract

A terminal device includes an acquisition unit which acquires local information of a present location at a present time; an accumulation unit which accumulates the acquired local information for a predetermined period; a communication unit which transmits the local information which is accumulated for a predetermined period to an information processing device, and receives parameters of a statistical model which are learned using the local information, which is acquired using the information processing device from a plurality of portable terminal devices, from the information processing device; and a prediction unit which predicts the local information in relation to an arbitrary time and location using the received parameters of the statistical model.

Claims

1. A terminal device comprising:

an acquisition unit which acquires local information of a present location at a present time;

an accumulation unit which accumulates the acquired local information for a predetermined period;

a communication unit which transmits the local information which is accumulated for a predetermined period to an information processing device, and receives parameters of a statistical model which are learned using the local information, which is acquired using the information processing device from a plurality of terminal devices, from the information processing device; and

a prediction unit which predicts the local information in relation to an arbitrary time and location using the received parameters of the statistical model.

2. The terminal device according to claim 1,

wherein the statistical model is a regression model in which at least a time and an area are explanatory variables, and items of local information are criterion variables.

3. The terminal device according to claim 2,

wherein, the communication unit receives only the parameters in which the weighting parameter is not zero, of weighting parameters of the regression model.

4. The terminal device according to claim 1,

wherein the local information is information representing a congestion factor of a base station to which the terminal device is connected in order to perform communication.

5. The terminal device according to claim 1,

wherein the prediction unit predicts the local information in relation to a time and a location specified by a user.

6. The terminal device according to claim 1,

wherein the prediction unit includes a function of predicting a movement route of the terminal device and a function of predicting a location and an arrival time of at least one point on the movement route, and predicts the local information in relation to the predicted location and arrival time.

7. The terminal device according to claim 1, further comprising:

a sensor which acquires predetermined data,

wherein the local information is the data acquired by the sensor.

8. The terminal device according to claim 1,

wherein the local information is evaluation information in relation to a predetermined index which is evaluated and input by a user that has the terminal device.

9. A terminal control method comprising:

causing a terminal device to acquire local information of a present location at a present time;

causing a terminal device to accumulate the acquired local information for a predetermined period;

causing a terminal device to transmit the local information, which is accumulated for a predetermined period to an information processing device, and to receive parameters of a statistical model which are learned using the local information which is acquired using the information processing device from a plurality of terminal devices, from the information processing device; and

causing a terminal device to predict the local information in relation to an arbitrary time and location using the received parameters of the statistical model.

10. A program causing a computer to function as:

an acquisition unit which acquires local information of a present location at a present time;

an accumulation unit which accumulates the acquired local information for a predetermined period;

a communication unit which transmits the local information which is accumulated for a predetermined period to an information processing device, and receives parameters of a statistical model which are learned using the local information, which is acquired using the information processing device from a plurality of terminal devices, from the information processing device; and

a prediction unit which predicts the local information in relation to an arbitrary time and location using the received parameters of the statistical model.

11. An information processing system formed from a terminal device and an information processing device,

wherein the terminal device includes

an acquisition unit which acquires local information of a present location at a present time;

an accumulation unit which accumulates the acquired local information for a predetermined period; and

a communication unit which transmits the local information which is accumulated for a predetermined period to the information processing device, and receives parameters of a statistical model which are learned using the local information, which is acquired using the information processing device from a plurality of terminal devices, from the information processing device,

wherein the information processing device includes

a communication unit which receives the local information for a predetermined period transmitted thereto from the terminal device, and transmits the parameters of the statistical model to the terminal device; and

a learning unit which learns the parameters of the statistical model using the local information received from a plurality of terminal devices, and

wherein either one of the terminal device or the information processing device includes

a prediction unit which predicts the local information in relation to an arbitrary time and location using the parameters of the statistical model.