Create Dataset For Activity Recognition

In this work, we introduce a new dataset, called Position-Aware Multi-Sensor (PAMS). The dataset contains both the accelerometer data and the gyroscope. The gyroscope data boosts the accuracy of activity recognition methods as well as enabling them to detect a wider range of activities. We also take the users’ information into account. Based on the biometric attributes of the participants, a separate learned model is generated to analyze their activities. We concentrated on several major activities, including sitting, standing, walking, running, ascending/descending stairs, and cycling. To evaluate the dataset, we use various classifiers, and the results are compared to WISDM. The outcome showed that the average precisions of all the activities are above 88.5% using the aforementioned classifiers.Now you can easly use thes codes to create your own dataset :)