Improving Text Classification with Vectors of Reduced Precision

Krzysztof Wróbel, Maciej Wielgosz, Marcin Pietron, Michal Karwatowski, Jerzy Duda, Aleksander Smywinski-Pohl

2018

Abstract

This paper presents the analysis of the impact of a floating-point number precision reduction on the quality of text classification. The precision reduction of the vectors representing the data (e.g. TF–IDF representation in our case) allows for a decrease of computing time and memory footprint on dedicated hardware platforms. The impact of precision reduction on the classification quality was performed on 5 corpora, using 4 different classifiers. Also, dimensionality reduction was taken into account. Results indicate that the precision reduction improves classification accuracy for most cases (up to 25% of error reduction). In general, the reduction from 64 to 4 bits gives the best scores and ensures that the results will not be worse than with the full floating-point representation.

Download


Paper Citation


in Harvard Style

Wróbel K., Wielgosz M., Pietron M., Karwatowski M., Duda J. and Smywinski-Pohl A. (2018). Improving Text Classification with Vectors of Reduced Precision.In Proceedings of the 10th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART, ISBN 978-989-758-275-2, pages 531-538. DOI: 10.5220/0006641505310538


in Bibtex Style

@conference{icaart18,
author={Krzysztof Wróbel and Maciej Wielgosz and Marcin Pietron and Michal Karwatowski and Jerzy Duda and Aleksander Smywinski-Pohl},
title={Improving Text Classification with Vectors of Reduced Precision},
booktitle={Proceedings of the 10th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,},
year={2018},
pages={531-538},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0006641505310538},
isbn={978-989-758-275-2},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 10th International Conference on Agents and Artificial Intelligence - Volume 2: ICAART,
TI - Improving Text Classification with Vectors of Reduced Precision
SN - 978-989-758-275-2
AU - Wróbel K.
AU - Wielgosz M.
AU - Pietron M.
AU - Karwatowski M.
AU - Duda J.
AU - Smywinski-Pohl A.
PY - 2018
SP - 531
EP - 538
DO - 10.5220/0006641505310538