Tufts ML Alumni
Associated Publications: [+]
Authors: Wallace, B., Dahabreh, I., Moran, K., Brodley, C. E., and Trikalinos, T.
KDD 2013 - Workshop on Data Mining
for Healthcare (DMH)
Current Research Topics:
Past Research Topics: [+]
Description: We are working with the Department of Geography at Boston University on the problem of defining the classes for land cover classification. Our method used both the unsupervised data and labels created from a previous classification scheme.
This work is partly supported by NSF grant IIS-0803409
Description: This project develops machine learning methods to health informatics, specifically with the aim of reducing the (human) workload involved in conducting systematic reviews. The idea is to use machine learning algorithms to induce models that semi-automate the clinical evidence synthesis process, thereby reducing workload. This has led to advances in ML including active learning under class imbalance, multiple expert active learning, incorporating labeled features into the SVM optimization, and how to determine when to stop labeling data.