Characteristic point detection system, characteristic point detection method, and program
First Claim
1. A characteristic point detection system, comprising:
- a staying area detection unit which performs clustering based on distribution of data points contained in GPS log data of a user, determines an index showing a staying time of the user in a cluster based on the number of data points contained in each cluster, extracts one or more clusters based on the index, and forms a staying area of the user based on the extracted clusters, wherewith a cluster is extracted with higher probability as the staying time is longer;
a representative point extraction unit which extracts, one at a time, a representative point of the staying area from each of the extracted staying areas, and determines a score of each of the representative points based on density of the data points in each of the staying areas; and
a representative point ranking unit which outputs a list ranking each of the representative points based on the score.
1 Assignment
0 Petitions
Accused Products
Abstract
The characteristic point detection system of this invention includes: a staying area detection unit 1003 which performs clustering based on distribution of data points contained in GPS log data of a user, determines an index showing a staying time of the user in a cluster based on a number of data points contained in each cluster, extracts one or more clusters based on the index, and forms a staying area of the user based on the extracted clusters, wherewith a cluster is extracted with higher probability as the staying time is longer; a representative point extraction unit 1004 which extracts, one at a time, a representative point of the staying area from each of the extracted staying areas, and determines a score of each of the representative points based on density of the data points in each of the staying areas; and an output unit 1005 which outputs a list ranking each of the representative points based on the score.
-
Citations
13 Claims
-
1. A characteristic point detection system, comprising:
-
a staying area detection unit which performs clustering based on distribution of data points contained in GPS log data of a user, determines an index showing a staying time of the user in a cluster based on the number of data points contained in each cluster, extracts one or more clusters based on the index, and forms a staying area of the user based on the extracted clusters, wherewith a cluster is extracted with higher probability as the staying time is longer; a representative point extraction unit which extracts, one at a time, a representative point of the staying area from each of the extracted staying areas, and determines a score of each of the representative points based on density of the data points in each of the staying areas; and a representative point ranking unit which outputs a list ranking each of the representative points based on the score. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8)
-
-
9. A characteristic point detection method, comprising:
-
a step of performing clustering based on distribution of data points contained in GPS log data of a user, determining an index showing a staying time of the user in a cluster based on the number of data points contained in each cluster, extracting one or more clusters based on the index, and forming a staying area of the user based on the extracted clusters, wherewith a cluster is extracted with higher probability as the staying time is longer; a step of extracting, one at a time, a representative point of the staying area from each of the extracted staying areas, and determining a score of each of the representative points based on density of the data points in each of the staying areas; and a step of outputting a list ranking each of the representative points based on the score.
-
-
10. A non-transitory computer readable medium storing a program which causes a computer to perform operations comprising:
-
performing clustering based on distribution of data points contained in GPS log data of a user, determining an index showing a staying time of the user in a cluster based on the number of data points contained in each cluster, extracting one or more clusters based on the index, and forming a staying area of the user based on the extracted clusters, wherewith a cluster is extracted with higher probability as the staying time is longer; extracting, one at a time, a representative point of the staying area from each of the extracted staying areas, and determining a score of each of the representative points based on density of the data points in each of the staying areas; and outputting a list ranking each of the representative points based on the score.
-
-
11. A characteristic point detection system, comprising:
-
a staying area detection unit which performs clustering based on distribution of data points contained in GPS log data of a user, extracts one or more clusters based on the number of data points contained in each cluster, and forms a staying area of the user based on the extracted clusters; a representative point extraction unit which extracts a representative point of the staying area from each of the extracted staying areas, and determines a score of each of the representative points; and an output unit which outputs the representative point based on the score.
-
-
12. A characteristic point detection method, comprising:
-
a step of performing clustering based on distribution of data points contained in GPS log data of a user, extracting one or more clusters based on the number of data points contained in each cluster, and forming a staying area of the user based on the extracted clusters; a step of extracting a representative point of the staying area from each of the extracted staying areas, and determining a score of each of the representative points; and a step of outputting the representative point based on the score.
-
-
13. A non-transitory computer readable storage medium storing a program which causes a computer to perform operations comprising:
-
performing clustering based on distribution of data points contained in GPS log data of a user, extracting one or more clusters based on the number of data points contained in each cluster, and forming a staying area of the user based on the extracted clusters; extracting a representative point of the staying area from each of the extracted staying areas, and determining a score of each of the representative points; and outputting the representative point based on the score.
-
Specification