Exploring the effectiveness of machine learning models on livestock behavior classification using 9-axial inertial sensor
摘要
Understanding the animal behavior has always been challenging for humans. Fortunately, the recent developments in electronic sensors can be helpful in analyzing and understanding the animal behavior. This work aims to explore the potential of using 9-axial inertial sensor and machine learning algorithms to understand the animal behavior more accurately. This study uses a dataset collected from cows equipped with 9-axial inertial sensor on their collar. The collected data is analyzed using various machine learning algorithms to classify the cattle’s behavior in different classes including standing, sitting, eating, rumination, etc. The findings of this research have significant implications for health management and livestock management, as the accurate classification of behavior can assist cattle farmers in monitoring and managing their herds more efficiently. The study also highlights the potential of using technology to improve our understanding of animal behavior, which can have significant impact for animal welfare and conservation efforts.