Human activity recognition is the task of identifying a person’s movement from sensors in a wearable device, such as a smartphone, smartwatch, or a medical-grade device. A great method for this task is machine learning, which is the study of algorithms that learn and improve on their own with the help of massive amounts of useful data. These classification models can accurately classify activities with the time-series data from accelerometers and gyroscopes. A significant way to improve the accuracy of these machine learning models is preprocessing the data, essentially augmenting data to make the identification of each activity, or class, easier for the model. <br/>On this topic, this paper explains the design of SigNorm, a new web application which lets users conveniently transform time-series data and view the effects of those transformations in a code-free, browser-based user interface. The second and final section explains my take on a human activity recognition problem, which involves comparing a preprocessed dataset to an un-augmented one, and comparing the differences in accuracy using a one-dimensional convolutional neural network to make classifications.
The City of Phoenix Street Transportation Department partnered with the Rob and Melani Walton Sustainability Solutions Service at Arizona State University (ASU) and researchers from various ASU schools to evaluate the effectiveness, performance, and community perception of the new pavement coating. The data collection and analysis occurred across multiple neighborhoods and at varying times across days and/or months over the course of one year (July 15, 2020–July 14, 2021), allowing the team to study the impacts of the surface treatment under various weather conditions.