What question did this study set out to answer?

The research aims to create a new dataset for improving the evaluation of machine learning models in handwritten character recognition.

January 25, 2026Scientific DataOpen Access

The dataset for extending EMNIST evaluation

View Full Paper

Ask AI

Bookmark

Authors

Describes a dataset that improves evaluation of handwriting models, suggesting enhanced analysis methods.

The research aims to create a new dataset for improving the evaluation of machine learning models in handwritten character recognition.
Developed a novel dataset combined with existing NIST Databases.
Analyzed popular machine learning models trained on the EMNIST-letters dataset.
Discussed evaluation issues related to accuracy comparisons on test sets.
Proposed evaluation methods on new independently constructed data.
Identified limitations in current evaluation methods for state-of-the-art models.
Demonstrated the potential for deeper insights using the new dataset and evaluation process.
Included publicly available source codes and datasets for the research community.

Szymański et al. (2026) studied this question.

View Full Paper

Ask AI

Bookmark