Selecting the best set of features to train predictive models is known to be a challenging problem. A bottom-up greedy feature selection method was employed to reduce the redundancy, noise, and low representativity of the 264 graph-based signature features obtained to represent molecules.

This method starts with zero features, by considering each feature independently, adds them one by one in accordance to a machine learning model, and keeps only the set of features with the most prominent performance metric (e.g., Pearson’s correlation coefficient) on that particular model.

Note: The content above has been extracted from a research article, so it may not display correctly.



Q&A
Please log in to submit your questions online.
Your question will be posted on the Bio-101 website. We will send your questions to the authors of this protocol and Bio-protocol community members who are experienced with this method. you will be informed using the email address associated with your Bio-protocol account.



We use cookies on this site to enhance your user experience. By using our website, you are agreeing to allow the storage of cookies on your computer.