Can I read Evaluating Classifiers for Real-World Problems on EtoBox?

Evaluating Classifiers for Real-World Problems by Durgesh Kumar is a document available to read on EtoBox.

What is Evaluating Classifiers for Real-World Problems about?

This document summarizes a study that evaluated 179 classifiers from 17 families using 121 datasets. The study aimed to determine the best overall classifier and rank classifiers and families. The random forest was found to perform the best, achieving 94.1% of maximum accuracy on average. Support vector machines with Gaussian and polynomial kernels also performed well. Using many classifiers, families, and datasets helped avoid bias and allowed conclusions not dependent on a specific collection.

Author

Durgesh Kumar

Language

EN