System for Detecting Potential Lost Person based on Conditional Random Field

International Conference on Computing and Applied Informatics 2016

物理学;计算机科学

Kusuma, R.S.^1 ; Saptawati, G.A.P.^1

Schools of Electrical Engineering and Informatics, Bandung Institute of Technology, Bandung, Indonesia^1

关键词: Conditional random field;    GPS data;    Gps trackers;    Region information;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/801/1/012002/pdf
DOI  :  10.1088/1742-6596/801/1/012002

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Global Positioning System (GPS) technology has been used widely in transsportation industry to help company in managing taxis. The most popular GPS utilization for taxi company is to identify the position of taxis and monitor theirs the mobility. Nowdays, data collected from GPS tracker is combined with data from taxi meter are analyzed to provide region information regarding potential passengers. Zicheng Liao's proposed a system based on GPS taxi data to detect anomalous area/region which was then interpreted as region with to predict rare passengers. The system was developed based on conditional random field (CRF) method and features position, velocity, passenger loading information. Our research was aimed to develop tool based on GPS data to detect potential lost person. We motivated by Liao research and modified the algorithms and features of CRF. Our experiments showed that the system has precision of 98.86% and recall of 87.478%.