Human Activity Recognition using SVM, KNN & Logistic Regression

Rohini Bhattarai, Vipra Bohara, L.N.Balai

Body Acceleration, Human Activity Recognition, K-Nearest Neighbor, Machine Learning, Postural Transitions, Support Vector Machine, Wireless Body Area Network.

Nowadays, Human activity recognition (HAR) is a very active yet challenging and demanding area and it has most popular uses of machine learning algorithms. HAR has a very significant role in different fields  such as  health care, theft detection, work monitoring in an organization and detecting emergencies, sports, smart home-based, childcare, security or work safety, human computer interaction, video surveillance system, robotics, daily monitoring, wildlife observation, and other diverse areas.

However, identifying human activities and actions is challenging from video sequences or still images is a challenging task due to the complexity of activities, speed of action, dynamic recording, background clutter, partial occlusion, changes in scale, viewpoint, lighting, and appearance and diverse application areas. Besides that, all the actions and activities are performed in distinct situations and backgrounds. Many applications, including video surveillance systems, human-computer interaction, and robotics for human behavior characterization, require a multiple activity recognition system.

There is a lot of work done in HAR; finding a suitable algorithm and sensors for a certain application area is still challenging. Here by using its embedded accelerometer or gyroscope in the classification model, SVM, KNN, and Logistic regression is implemented. The purpose of this paper is, to solve the HAR problem under wireless body area network  to achieve the profitable HAR method.