DS-200 · Question #62
There are 20 patients with acute lymphoblastic leukemia (ALL) and 32 patients with acute myeloid leukemia (AML), both variants of a blood cancer. The makeup of the groups as follows: Each individual h
Sign in or unlock DS-200 to reveal the answer and full explanation for question #62. The question stem and answer options stay visible for context.
Question
There are 20 patients with acute lymphoblastic leukemia (ALL) and 32 patients with acute myeloid leukemia (AML), both variants of a blood cancer. The makeup of the groups as follows:
Each individual has an expression value for each of 10000 different genes. The expression value for each gene is a continuous value between -1 and 1. You've built your model for discriminating between AML and ALL patients and you find that it works quite well on your current data. One month later, a collaboration tells you she has fresh data from 100 new AML/ALL patients. You run the samples through your model, and turns out your model has very poor predictive accuracy on the new samples; specifically, your model predicts that all males have ALL. What is the most reliable way to fix this problem?
Options
- AChange the distance metric
- BReduce the number of dimensions
- CUse a Gibbs sampler on a Bayesian network
- DPerform matched sampling across other provided variables
Unlock DS-200 to see the answer
You've previewed enough free DS-200 questions. Unlock DS-200 for full answers, explanations, the timed quiz mode, progress tracking, and the master PDF. Question stem and options stay visible so you can still see what's on the exam.