Gabriel Mersy

Gabriel Mersy, Vincent Santore, Isaac Rand et al.

We present a novel approach to the measurement of American state legislature polarization with an experimental comparison of three different machine learning algorithms. Our approach strictly relies on public data sources and open source software. The results suggest that artificial neural network regression has the best outcome compared to both support vector machine and ordinary least squares regression in the prediction of both state House and state Senate legislature polarization. In addition to the technical outcomes of our study, broader implications are assessed as a means of highlighting the importance of accessible information for the higher purpose of promoting civic responsibility.

Gabriel Mersy, Jin Hong Kuan

Given recent advances in deep music source separation, we propose a feature representation method that combines source separation with a state-of-the-art representation learning technique that is suitably repurposed for computer audition (i.e. machine listening). We train a depthwise separable convolutional neural network on a challenging electronic dance music (EDM) data set and compare its performance to convolutional neural networks operating on both source separated and standard spectrograms. It is shown that source separation improves classification performance in a limited-data setting compared to the standard single spectrogram approach.