Abstract:
The difficulty of identifying a body's behavior based on sensor data is known as activity recognition. It's among the most widely studied topics in the field of machine learning-based classification. Bento Packing Activity Recognition Challenge (BPARC) asked participants to recognize bento packing preparation using motion capture sensors. Thirteen motion capture body markers were used to capture the BPARC dataset. The markers were three axes (x, y, and z) with a total of thirty-nine axes. One of the most challenging difficulties to solve in this investigation was identifying complicated tasks as smaller activities that are part of larger activities. Using Random Forest, Support Vector Machine, and Logistic Regression, we’ve built a machine learning approach that extracts dynamical data for bento packing activity identification. The model we proposed for that kind of dataset has a classification accuracy of 76%, 62% and 63% for RF, SVM and LR, respectively.
