III: Small: Exploring Data in Multiple Clustering Views
The primary objective of this research is to formulate a framework for a new paradigm for clustering: discovering all possible non-redundant multiple clustering views from data. Typical clustering algorithms only find one clustering solution, but many real and complex data are multi-faceted by nature. Data can be interpreted in many different ways. Given the same data, what is interesting to a physician will be different from what is important to an insurance agency. This research will provide new formulations, algorithms, and tools for exploratory data analysis that are widely applicable to many domains. The PI will apply the new algorithms in detection of skin lesions, including cancers/ This will involve the automatic segmentation of the dermis and epidermis junction in skin images, automated detection of machine sounds, and in developing algorithms for multiple non-redundant clustering of text and natural images. Additionally, this project will provide research experiences for both undergraduate and graduate students in the classroom and in the lab. The PI will work with the Society of Women Engineers to inspire female students to pursue careers in engineering and computer science, and to ensure that under-represented groups are involved in this research.
Northeastern University’s College of Engineering is home to numerous federally-funded research centers and an array of leading-edge projects and initiatives that advance discovery and new knowledge in health, sustainability, and security.