SVM-RBF Kernel Learning Model for Activity Recognition in Smart Home

被引：1

作者：

Chou, Zhi-Wei ^{[1
]}

Lu, Ying-Kai ^{[1
]}

Huang, Ke-Nung ^{[1
]}

机构：

[1] I Shou Univ, Dept Elect Engn, Kaohsiung, Taiwan

来源：

22ND IEEE/ACIS INTERNATIONAL CONFERENCE ON SOFTWARE ENGINEERING, ARTIFICIAL INTELLIGENCE, NETWORKING AND PARALLEL/DISTRIBUTED COMPUTING (SNPD 2021-FALL) | 2021年

关键词：

abnormal behavior; activity recognition; smart home;

D O I：

10.1109/SNPD51163.2021.9704919

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

The world population is aging. Taiwan will have at least 20 percent of the population over 65 by 2026. Telemonitoring technology is one of the solutions used to assist elderly people live independently. We designed a SVM-RBF kernel learning model to classify activities of daily living and to analyze an individual's daily routines and habits, typically for the elderly who live alone. One of the CASAS smart home datasets was used to train and to retest the algorithm. A non-trained dataset was also used to validate the accuracy of the algorithm. Abnormal behaviors can be detected by compared with individual's daily activity pattern as baseline.

引用

页码：130 / 133

页数：4