Filtering by
- Creators: Computer Science and Engineering Program
- Creators: Mazzola, Daniel
- Member of: Barrett, The Honors College Thesis/Creative Project Collection
- Member of: Theses and Dissertations
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.
Compass portal features tools that help teachers, psychologists, behavioral specialists gain insights on students’ performance through activities they have completed.
COMPASS portal features tools that help teachers, psychologists, behavioral Specialists gain insights on students’ performance through activities they have completed.
Anthemy is a web app that I created so that Spotify users could connect with other uses and see their listening statistics. The app has a chat feature that matches concurrent users based on a variety of search criteria, as well as a statistics page that contains a breakdown of a user's top artists, songs, albums, and genres as well as a detailed breakdown of each of their liked playlists.
Song Sift is an application built using Angular that allows users to filter and sort their song library to create specific playlists using the Spotify Web API. Utilizing the audio feature data that Spotify attaches to every song in their library, users can filter their downloaded Spotify songs based on four main attributes: (1) energy (how energetic a song sounds), (2) danceability (how danceable a song is), (3) valence (how happy a song sounds), and (4) loudness (average volume of a song). Once the user has created a playlist that fits their desired genre, he/she can easily export it to their Spotify account with the click of a button.