Comparison and Evaluation of Data Composition and Deep Learning Models in Archival Handwritten Digit Classification
Journal of Machine Intelligence and Data Science · 2022
DOI: 10.11159/jmids.2022.001 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Machine Intelligence and Data Science, 2022. Verfasst von Nathan LeBlanc und 1 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 20 Literaturangaben.
Abstract (Originalsprache)
– Archival maritime logs are well-preserved treasure throve of climate-related data. The analysis of these documents is instrumental to understanding historical climate trends and future predictions. Transcribing such handwritten logs depends on handwritten letter/digit recognition, which is our aim. The shortcomings of OCR (Optical Character Recognition) are manifesting in frequent confusion of digits and letters when it comes to archival handwritten documents. In this extension of conference and thesis work, two such methods are put to the test – convolutional (CNN) and long-short term memory (LSTM) neural networks (NN). A compound model of convolutional NN followed by LSTM is also considered. While all models register high accuracy, it is observed that the compound model performs faster with accuracy above the lone CNN. We also analyse dataset composition and test for size and balance.
Zitieren
APA
LeBlanc, N., & Valova, I. (2022). Comparison and Evaluation of Data Composition and Deep Learning Models in Archival Handwritten Digit Classification. Journal of Machine Intelligence and Data Science. https://doi.org/10.11159/jmids.2022.001
Vancouver
LeBlanc N, Valova I. Comparison and Evaluation of Data Composition and Deep Learning Models in Archival Handwritten Digit Classification. Journal of Machine Intelligence and Data Science. 2022. doi:10.11159/jmids.2022.001
BibTeX
@article{LeBlanc2022,
author = {LeBlanc, Nathan and Valova, Iren},
title = {Comparison and Evaluation of Data Composition and Deep Learning Models in Archival Handwritten Digit Classification},
journal = {Journal of Machine Intelligence and Data Science},
year = {2022},
doi = {10.11159/jmids.2022.001}
}
EUVIMED