*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: Using Modified Sparse Coding as an Unsupervised Feature Extractor for the Binary Classification of Imbalanced Datasets
Committee:
Dr. Anderson, Advisor
Dr. Rozell, Chair
Dr. Romberg
Abstract:
The objective of the proposed research is to explore the use of sparse coding as a tool for unsupervised feature learning to help in the classification of imbalanced datasets. Traditional sparse coding dictionaries are learned by minimizing the average approximation error between a vector and its sparse decomposition. As such, these dictionaries may overlook important features that are only present in the minority class. Without these features, it may be difficult to accurately classify between the two classes. To overcome this problem, this proposed work will explore novel modifications to the dictionary learning framework that will encourage dictionaries to include anomalous features. Sparse coding also inherently assumes that a vector can be represented as a sparse linear combination of a feature set. This proposal will address the ability of sparse coding to learn a representative dictionary when the underlying data has a nonlinear sparse structure. Finally, this work will apply the intuition gained from exploring dictionary learning in this manner to the binary classification of imbalanced datasets.
