Behaviour Pattern Analysis System, Mobile Terminal, Behaviour Pattern Analysis Method, and Program

• US 20110081634A1
• Filed: 08/25/2010
• Published: 04/07/2011
• Est. Priority Date: 10/02/2009
• Status: Active Grant

• Current Assignee: Sony Corporation (Sony Group Corp.)
• Original Assignee: Sony Corporation (Sony Group Corp.)
• Inventors: Fukumoto, Yasutaka, Katsu, Masanori, Ogata, Takashi, Kurata, Masatomo, Murata, Makoto
• Granted Patent: US 8,438,127 B2
• Time in Patent Office: Days
• US Class Current: 434/236
• CPC Class Codes: G01C 21/20 Instruments for performing navigational calculations G01C21/24, G01C21/26 take precedence
• CPC Class Codes: G06F 3/011 Arrangements for interaction with the human body, e.g. for user immersion in virtual reality blind teaching G09B21/00
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06Q 10/00 Administration; Management
• CPC Class Codes: H04L 67/306 User profiles
• CPC Class Codes: H04W 4/021 Services related to particular areas, e.g. point of interest [POI] services, venue services or geofences
• CPC Class Codes: H04W 4/026 using orientation information, e.g. compass
• CPC Class Codes: H04W 4/027 using movement velocity, acceleration information
• CPC Class Codes: H04W 4/029 Location-based management or tracking services
• CPC Class Codes: H04W 4/33 for indoor environments, e.g. buildings

Abstract

Provided is a mobile terminal including a movement sensor that detects a movement of a user and outputs movement information, acquires information on a building existing at a current location or information on buildings existing in a vicinity of the current location, analyses the movement information output from the movement sensor, and detects a first behaviour pattern corresponding to the movement information from multiple first behaviour patterns obtained by classifying behaviours performed by the user over a relatively short period of time, and analyses the information on a building or buildings and the first behaviour pattern, and detects a second behaviour pattern corresponding to the information on a building or buildings and the first behaviour pattern from multiple second behaviour patterns obtained by classifying behaviours performed by the user over a relatively long period of time.

Claims

1. A behaviour pattern analysis system comprising:

a mobile terminal including

a movement sensor that detects a movement of a user and outputs movement information,

a current location information acquisition unit that acquires information on a current location,

a building information acquisition unit that acquires information on a building existing at a location indicated by the information acquired by the current location information acquisition unit or information on buildings existing at the current location and in a vicinity of the current location,

a first behaviour pattern detection unit that analyses the movement information output from the movement sensor, and detects a first behaviour pattern corresponding to the movement information from multiple first behaviour patterns obtained by classifying behaviours performed by the user over a relatively short period of time, and

a transmission unit that transmits, to a server, the information on a building or buildings acquired by the building information acquisition unit and the first behaviour pattern detected by the first behaviour pattern detection unit; and

a server including

a reception unit that receives, from the mobile terminal, the information on a building or buildings and the first behaviour pattern, and

a second behaviour pattern detection unit that analyses the information on a building or buildings and the first behaviour pattern received by the reception unit, and detects a second behaviour pattern corresponding to the information on a building or buildings and the first behaviour pattern from multiple second behaviour patterns obtained by classifying behaviours performed by the user over a relatively long period of time.

2. The behaviour pattern analysis system according to claim 1, wherein the second behaviour pattern detection unit creates, by using a specific machine learning algorithm, a detection model for detecting the second behaviour pattern from the information on a building or buildings and the first behaviour pattern, and detects, by using the created detection model, the second behaviour pattern corresponding to the information on a building or buildings and the first behaviour pattern received by the reception unit.

3. The behaviour pattern analysis system according to claim 1,

wherein the mobile terminal further includes a time information acquisition unit that acquires information on a time of a time point of acquisition of the information on a current location by the current location information acquisition unit,

wherein the transmission unit transmits, to the server, the information on a building or buildings acquired by the building information acquisition unit, the first behaviour pattern detected by the first behaviour pattern detection unit and the information on a time acquired by the time information acquisition unit,

wherein the server holds, for each combination of the first behaviour pattern and the information on a time, a score map assigning a score to each combination of the information on a building or buildings and the second behaviour pattern,

wherein, in a case the score map is selected based on the first behaviour pattern detected by the first behaviour pattern detection unit and the information on a time acquired by the time information acquisition unit, a combination of scores corresponding to the information on a building existing at the current location acquired by the building information acquisition unit is extracted from the selected score map and a highest score in the extracted combination of scores is a specific value or less, the second behaviour pattern detection unit creates, by using a specific machine learning algorithm, a detection model for detecting the second behaviour pattern from the information on a building or buildings and the first behaviour pattern, and detects, by using the created detection model, the second behaviour pattern corresponding to the information on a building or buildings and the first behaviour pattern received by the reception unit.

4. The behaviour pattern analysis system according to claim 1,

wherein the mobile terminal further includes

a storage unit in which schedule information recording, in a time-series manner, a behaviour of a user capable of being expressed by a combination of the second behaviour patterns is stored,

a matching determination unit that reads the schedule information stored in the storage unit, and determines whether a present behaviour, a past behaviour and a future behaviour of the user recorded in the schedule information and the second behaviour pattern detected by the second behaviour pattern detection unit match or not, and

a display unit that displays, according to a result of determination by the matching determination unit, whether an actual behaviour matches a schedule recorded in the schedule information, is behind the schedule, or is ahead of the schedule.

5. The behaviour pattern analysis system according to claim 4,

wherein the server further includes a behaviour prediction unit that predicts, by using a history of the second behaviour patterns detected by the second behaviour pattern detection unit, a second behaviour pattern to be performed by the user next,

wherein, in a case of determining that a behaviour of the user matching the second behaviour pattern is not recorded in the schedule information, the matching determination unit acquires, from the server, the second behaviour pattern predicted by the behaviour prediction unit and extracts, from the schedule information, a behaviour of the user matching the acquired second behaviour pattern, and

wherein the display unit displays information relating to the behaviour of the user extracted by the matching determination unit.

6. The behaviour pattern analysis system according to claim 5,

wherein the server includes

a location information accumulation unit that receives, by the reception unit, the information on a current location acquired by the current location information acquisition unit and the first behaviour pattern detected by the first behaviour pattern detection unit, and stores the information on a current location and a history of the first behaviour patterns in the storage unit, and

a clustering unit that clusters places where the user stays for a long time, by using the information on a current location and the history of the first behaviour patterns accumulated in the storage unit by the location information accumulation unit, and calculates a staying probability of staying at each of the places and a movement probability of moving between the places, and

wherein the behaviour prediction unit predicts the second behaviour pattern to be performed by the user next, based on the staying probability and the movement probability calculated by the clustering unit.

7. The behaviour pattern analysis system according to claim 1,

wherein the behaviour pattern analysis system includes multiple mobile terminals, and

wherein the server further includes

a notification information storage unit that stores, in association with each other, notification information to be notified at a specific time and a specific second behaviour pattern, and

an information notification unit that, at the specific time, refers to the second behaviour pattern detected by the second behaviour pattern detection unit based on the information on a building or buildings and the first behaviour pattern received by the reception unit from each of the mobile terminals, and notifies a mobile terminal corresponding to a second behaviour pattern same as the specific second behaviour pattern of the notification information.

8. The behaviour pattern analysis system according to claim 7, wherein the information notification unit counts the number of mobile terminals corresponding to a second behaviour pattern same as the specific second behaviour pattern, and in a case the number of the mobile terminals is a specific number or more, notifies all of the multiple mobile terminals of the notification information.

9. A mobile terminal comprising:

a movement sensor that detects a movement of a user and outputs movement information;

a current location information acquisition unit that acquires information on a current location;

a building information acquisition unit that acquires information on a building existing at a location indicated by the information acquired by the current location information acquisition unit or information on buildings existing at the current location and in a vicinity of the current location;

a first behaviour pattern detection unit that analyses the movement information output from the movement sensor, and detects a first behaviour pattern corresponding to the movement information from multiple first behaviour patterns obtained by classifying behaviours performed by the user over a relatively short period of time; and

a second behaviour pattern detection unit that analyses the information on a building or buildings acquired by the building information acquisition unit and the first behaviour pattern detected by the first behaviour pattern detection unit, and detects a second behaviour pattern corresponding to the information on a building or buildings and the first behaviour pattern from multiple second behaviour patterns obtained by classifying behaviours performed by the user over a relatively long period of time.

10. The mobile terminal according to claim 9, further comprising:

a time information acquisition unit that acquires information on a time of a time point of acquisition of the information of a current location by the current location information acquisition unit,

wherein a score map assigning a score to each combination of the information on a building or buildings and the second behaviour pattern is provided for each combination of the first behaviour pattern and the information on a time, and

wherein the second behaviour pattern detection unit

selects the score map based on the first behaviour pattern detected by the first behaviour pattern detection unit and the information on a time acquired by the time information acquisition unit,

extracts, from the selected score map, a combination of scores corresponding to the information on a building existing at the current location acquired by the building information acquisition unit, and

detects the second behaviour pattern corresponding to a highest score in the extracted combination of scores.

11. The mobile terminal according to claim 9, further comprising:

a time information acquisition unit that acquires information on a time of a time point of acquisition of the information of a current location by the current location information acquisition unit,

wherein a score map assigning a score to each combination of the information on a building or buildings and the second behaviour pattern is provided for each combination of the first behaviour pattern and the information on a time,

wherein the building information acquisition unit acquires, as the information on buildings existing at the current location and in a vicinity of the current location, category types of the buildings and the number of buildings for each category type, and

wherein the second behaviour pattern detection unit

selects the score map based on the first behaviour pattern detected by the first behaviour pattern detection unit and the information on a time acquired by the time information acquisition unit,

extracts, from the selected score map, combinations of scores corresponding to respective category types acquired by the building information acquisition unit,

normalises, by respective highest scores, each score included in the combinations of scores corresponding to respective category types,

performs weighting on the normalised combinations of scores corresponding to respective category types according to the number of buildings for each category type acquired by the building information acquisition unit, and

adds, for each second behaviour pattern, the weighted scores corresponding to the respective category types, and detects the second behaviour pattern for which a result of addition is greatest.

12. The mobile terminal according to claim 9, further comprising:

a display unit on which a display object for starting an application associated with the second behaviour pattern is displayed; and

a display control unit that makes the display unit preferentially display, according to the second behaviour pattern detected by the second behaviour pattern detection unit, the display object associated with the second behaviour pattern.

13. A behaviour pattern analysis server comprising:

a reception unit that receives, from a mobile terminal including a movement sensor that detects a movement of a user and outputs movement information and a current location information acquisition unit that acquires information on a current location, the movement information and the information on a current location;

a building information acquisition unit that acquires information on a building existing at a location indicated by the information on a current location received by the reception unit or information on buildings existing at the current location and in a vicinity of the current location;

a first behaviour pattern detection unit that analyses the movement information received by the reception unit, and detects a first behaviour pattern corresponding to the movement information from multiple first behaviour patterns obtained by classifying behaviours performed by the user over a relatively short period of time; and

a second behaviour pattern detection unit that analyses the information on a building or buildings acquired by the building information acquisition unit and the first behaviour pattern detected by the first behaviour pattern detection unit, and detects a second behaviour pattern corresponding to the information on a building or buildings and the first behaviour pattern from multiple second behaviour patterns obtained by classifying behaviours performed by the user over a relatively long period of time.

14. A behaviour pattern analysis method comprising the steps of:

acquiring movement information indicating a result of detection by a movement sensor for detecting a movement of a user;

acquiring information on a current location;

acquiring information on a building existing at a location indicated by the information on a current location acquired in the step of acquiring information on a current location or information on buildings existing at the current location and in a vicinity of the current location;

analysing the movement information acquired in the step of acquiring movement information, and detecting a first behaviour pattern corresponding to the movement information from multiple first behaviour patterns obtained by classifying behaviours performed by the user over a relatively short period of time; and

analysing the information on a building or buildings acquired in the step of acquiring information on a building or buildings and the first behaviour pattern detected in the step of analysing the movement information and detecting a first behaviour pattern, and detecting a second behaviour pattern corresponding to the information on a building or buildings and the first behaviour pattern from multiple second behaviour patterns obtained by classifying behaviours performed by the user over a relatively long period of time.

15. A program for causing a computer to realise:

a movement information acquisition function of acquiring movement information indicating a result of detection by a movement sensor for detecting a movement of a user;

a current location information acquisition function of acquiring information on a current location;

a building information acquisition function of acquiring information on a building existing at the current location indicated by the information acquired by the current location information acquisition function or information on buildings existing at the current location and in a vicinity of the current location;

a first behaviour pattern detection function of analysing the movement information acquired by the movement information acquisition function, and detecting a first behaviour pattern corresponding to the movement information from multiple first behaviour patterns obtained by classifying behaviours performed by the user over a relatively short period of time; and

a second behaviour pattern detection function of analysing the information on a building or buildings acquired by the building information acquisition function and the first behaviour pattern detected by the first behaviour pattern detection function, and detecting a second behaviour pattern corresponding to the information on a building or buildings and the first behaviour pattern from multiple second behaviour patterns obtained by classifying behaviours performed by the user over a relatively long period of time.