In this project, we investigated novel computational methods for discovering useful patterns and knowledge from the data sets of two specific domains: 1. The Web of Science mentions in scientific publications and 2. Physical activity data of COPD patients captured by monitoring devices. For the first data set, we explored the application of cutting edge machine learning and natural language processing (NLP) techniques to a subset of both meta-data and full-text scientific publications in hopes of finding new ways for users to derive actionable insights from this content. For the second data set, we applied a variety of statistical and machine learning techniques for visualizing, modeling, and predicting patients’ behaviors. A visualization tool has been developed for the first data set and meaningful and useful analytic results have been reported back to the IAB mentor of the second data set.