A counterintuitive approach yields big benefits for high-dimensional, small-sized dataset problems

Extracting meaningful information out of clinical datasets can mean the difference between a successful diagnosis and a protracted illness. However datasets can vary widely both in terms of the number of 'features' measured and the number of independent observations taken. Now, A*STAR researchers have developed an approach for targeted feature selection from datasets with small sample sizes, which tackles the so-called class imbalance problem.

Read the whole article on Physorg

More from Physorg