TY - CHAP
T1 - An optimization approach for elementary school handwritten mathematical expression recognition
AU - Chevtchenko, Sergio F.
AU - Carvalho, Ruan
AU - Rodrigues, Luiz
AU - Souza, Everton
AU - Rosa, Daniel
AU - Cordeiro, Filipe
AU - Pereira, Cicero
AU - Vieira, Thales
AU - Marinho, Marcelo
AU - Dermeval, Diego
AU - Bittencourt, Ig Ibert
AU - Isotani, Seiji
AU - Macario, Valmir
PY - 2024
Y1 - 2024
N2 - This study introduces a novel approach to Handwritten Mathematical Expression Recognition (HMER), focusing on elementary school mathematical expressions. Recognizing the challenges posed by limited training data and the unique characteristics of elementary students' handwriting, we present a multiobjective optimization method tailored for small training datasets. We employ state-of-the-art HMER methods, including transformer-based and attention mechanism models, and optimize them using a custom dataset comprised of elementary school arithmetic equations. This dataset contains 1237 images and includes both horizontal and vertical equations and isolated numbers, featuring common errors in children's handwriting. Additional similar datasets are also leveraged for training augmentation. Our experimental results demonstrate the efficacy of the optimization approach, significantly improving the performance of the evaluated models in terms of expression recognition rate and inference speed. This study contributes to the field of HMER by providing an effective optimization approach for SOTA models and by introducing a specialized dataset for elementary school mathematics. The dataset is available upon request.
AB - This study introduces a novel approach to Handwritten Mathematical Expression Recognition (HMER), focusing on elementary school mathematical expressions. Recognizing the challenges posed by limited training data and the unique characteristics of elementary students' handwriting, we present a multiobjective optimization method tailored for small training datasets. We employ state-of-the-art HMER methods, including transformer-based and attention mechanism models, and optimize them using a custom dataset comprised of elementary school arithmetic equations. This dataset contains 1237 images and includes both horizontal and vertical equations and isolated numbers, featuring common errors in children's handwriting. Additional similar datasets are also leveraged for training augmentation. Our experimental results demonstrate the efficacy of the optimization approach, significantly improving the performance of the evaluated models in terms of expression recognition rate and inference speed. This study contributes to the field of HMER by providing an effective optimization approach for SOTA models and by introducing a specialized dataset for elementary school mathematics. The dataset is available upon request.
KW - Deep Learning
KW - HMER
KW - Optimization
UR - https://www.scopus.com/pages/publications/85199804989
UR - https://go.openathens.net/redirector/westernsydney.edu.au?url=https://doi.org/10.1007/978-3-031-64312-5_28
U2 - 10.1007/978-3-031-64312-5_28
DO - 10.1007/978-3-031-64312-5_28
M3 - Chapter
AN - SCOPUS:85199804989
SN - 9783031643118
T3 - Communications in Computer and Information Science
SP - 234
EP - 241
BT - Artificial Intelligence in Education, Posters and Late Breaking Results, Workshops and Tutorials, Industry and Innovation Tracks, Practitioners, Doctoral Consortium and Blue Sky, 25th International Conference, AIED 2024, Recife, Brazil, July 8-12, 2024, Proceedings, Part II
A2 - Olney, Andrew M.
A2 - Chounta, Irene-Angelica
A2 - Liu, Zitao
A2 - Santos, Olga C.
A2 - Bittencourt, Ig Ibert
PB - Springer
CY - Switzerland
ER -