Article ID | Journal | Published Year | Pages | File Type |
---|---|---|---|---|
6965295 | Accident Analysis & Prevention | 2018 | 13 Pages |
Abstract
Increasing the size of training data alone cannot solve the problem of poor classification performance on rare categories by ML models. Filtering could be an effective strategy to improve classification performance of rare categories when large training data is not available.
Related Topics
Physical Sciences and Engineering
Chemical Engineering
Chemical Health and Safety
Authors
Gaurav Nanda, Kirsten Vallmuur, Mark Lehto,