JIITA, vol.3 no.2, pp.263-266, 2019, DOI: 10.22664/ISITA.2019.3.2.263
SANGMIN KIM 1), SANGWOOK PARK 2), YEONGSU YU 2), HENGSUK LEE 1,*)
1) System Technology Center. GIVET, Yeongcheon-si, Korea
2) Institute of Technology. Daedong Tech, Kyungsan-si, Korea
Abstract: The purpose of this study is to find and classify the determinants of cattle’s behavior patterns. Many researchers have studied similar researches as this topic. However, the accuracy of the data analysis is still insufficient. In this paper, we attempt to improve the accuracy of existing classification methods by analyzing the data of previous researchers and proposing a classification method of new cattle’s behavior patterns. To improve the accuracy of cattle’s behavior patterns that we are tried to extract the frequency and value of the each signal by Fast Fourier Transform, and then we input the frequency and value to classify cattle’s behavior patterns at the designed Support Vector Machine. We can get high accuracy of classification of the cattle’s behavior patterns.
Keyword: Cattle’s Behavior Patterns; insert livestock health care; smart farm; Support Vector Machine;
Fullpaper: