| dc.description.abstract | Since 1990s, activity recognition effectual field in machine learning literature. Most of studies that relevant activity recognition, use feature extraction method to achieve higher classification performance. Moreover, these studies mostly use traditional machine learning algorithms for classification. In this paper, we focus on a deep (Long Short Term Memory) LSTM neural network for feature free classification of seven daily activities by using raw data that collected from three-dimensional accelerometer. Based on the results, the proposed deep LSTM approach can classify raw data with high performance. The results show that the proposed deep LSTM approach achieved 91.34, 96.91, 88.78, 87.58 as percent classification performance in terms of accuracy, sensitivity, specificity, F-measure respectively. | en_US |
| dc.description.sponsorship | IEEE Reg 8; IEEE Hungary Sect; IEEE Czechoslovakia Sect & SP CAS COM Joint Chapter; Sci Assoc Infocommunicat; Brno Univ Technol, Dept Telecommunicat; Budapest Univ Technol & Econ, Dept Telecommunicat & Media Informat; Czech Tech Univ Prague, Dept Telecommunicat Engn; Isik Univ, Dept Elect & Elect Engn,; Istanbul Tech Univ, Elect & Communicat Engn Dept; Josip Juraj Strossmayer Univ Osijek, Fac Elect Engn Comp Sci & Informat Technol; Karadeniz Tech Univ, Dept Elect & Elect Engn; Natl Taiwan Univ Sci & Technol, Dept Elect & Comp Engn; Seikei Univ, Grad Sch & Fac Sci & Technol, Informat Networking Lab; Slovak Univ Technol Bratislava, Inst Multimedia Informat & Commun Technologies; Escola Univ Politecnica Mataro, Tecnocampus; Tech Univ Sofia, Fac Telecommunicat; Univ Paris 8, UFR MITSIC, Lab Informatique Avancee Saint Denis; Univ Politehnica Bucharest, Ctr Adv Res New Mat Prod & Innovat Proc; Univ Ljubljana, Lab Telecommunicat; Univ Patras, Phys Dept; VSB Tech Univ Ostrava, Dept Telecommunicat; W Pomeranian Univ Technol, Fac Elect Engn | en_US |
