A Technical Feasibility Study of Handwritten Digit Recognition Using VGG19, XGBoost, and Structural Similarity Analysis for Digital Learning Systems

Monika Mehta; Ashmeet Kaur; Madhulika Bhatia

doi:10.14710/jmasif.17.1.78073

DOI: https://doi.org/10.14710/jmasif.17.1.78073

A Technical Feasibility Study of Handwritten Digit Recognition Using VGG19, XGBoost, and Structural Similarity Analysis for Digital Learning Systems

Monika Mehta¹

, Ashmeet Kaur²

, Madhulika Bhatia³

¹Department of Fashion Communication, National Institute of Fashion Technology, India

²Engineer 1, Numerator, India

³Software Development Officer, Digital Health and Care Wales, United Kingdom

Received: 25 Sep 2025; Revised: 1 Apr 2026; Accepted: 16 Apr 2026; Available online: 20 Apr 2026; Published: 6 May 2026.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

BibTex Citation Data :

@article{JMASIF78073,
    author = {Monika Mehta and Ashmeet Kaur and Madhulika Bhatia},
    title = {A Technical Feasibility Study of Handwritten Digit Recognition Using VGG19, XGBoost, and Structural Similarity Analysis for Digital Learning Systems},
    journal = {Jurnal Masyarakat Informatika},
  volume = {17},
    number = {1},
    year = {2026},
    keywords = {Handwritten Digit Recognition; Transfer Learning; VGG19; XG Boost; Structural Similarity Index; Educational Technology Feasibility},
    abstract = {  Handwritten digit recognition plays an important role in intelligent educational interfaces, early learning analytics, and human–computer interaction. This study investigates the technical feasibility of integrating transfer learning and gradient boosting for low-resource handwritten digit recognition environments relevant to emerging digital education platforms. A pre-trained VGG19 network is employed as a deep feature extractor, followed by classification using XGBoost. Structural Similarity Index (SSIM) analysis is incorporated as a feedback metric to quantify visual similarity between learner input and reference digit forms. The work is positioned as a feasibility validation of a hybrid vision-learning pipeline under limited-data conditions to explore its applicability in future AI-assisted handwriting support systems. Experiments are conducted using controlled augmentation, k-fold cross-validation, and comparative baseline CNN models. Results demonstrate stable classification behaviour and consistent SSIM-based similarity scoring, supporting the suitability of the architecture as a building block for adaptive digital handwriting evaluation systems. The framework demonstrates potential for integration into next-generation digital learning tools, including early education, special education support, skill assessment interfaces, and AI-driven formative feedback platforms.  },
   issn = {2777-0648},  doi = {10.14710/jmasif.17.1.78073},
    url = {https://ejournal.undip.ac.id/index.php/jmasif/article/view/78073}
}

Citation Format:

Abstract

Handwritten digit recognition plays an important role in intelligent educational interfaces, early learning analytics, and human–computer interaction. This study investigates the technical feasibility of integrating transfer learning and gradient boosting for low-resource handwritten digit recognition environments relevant to emerging digital education platforms. A pre-trained VGG19 network is employed as a deep feature extractor, followed by classification using XGBoost. Structural Similarity Index (SSIM) analysis is incorporated as a feedback metric to quantify visual similarity between learner input and reference digit forms. The work is positioned as a feasibility validation of a hybrid vision-learning pipeline under limited-data conditions to explore its applicability in future AI-assisted handwriting support systems. Experiments are conducted using controlled augmentation, k-fold cross-validation, and comparative baseline CNN models. Results demonstrate stable classification behaviour and consistent SSIM-based similarity scoring, supporting the suitability of the architecture as a building block for adaptive digital handwriting evaluation systems. The framework demonstrates potential for integration into next-generation digital learning tools, including early education, special education support, skill assessment interfaces, and AI-driven formative feedback platforms.

Fulltext Email colleagues

Keywords: Handwritten Digit Recognition; Transfer Learning; VGG19; XG Boost; Structural Similarity Index; Educational Technology Feasibility

Article Metrics:

Article Info

Section: Research Article

Language : EN

In Vol 17, No 1 (2026): May 2026 (Ongoing)

Most cited articles

Enterprise Architecture Model untuk Aplikasi Government Studi Komparasi Ekstraksi Fitur pada Pengenalan Wajah Menggunakan Principal Component Analysis (PCA) dan Wavelet Daubechies Pengenalan Jenis Golongan Darah Menggunakan Jaringan Syaraf Tiruan Perceptron APLIKASI SISTEM INFORMASI GEOGRAFIS BERBASIS WEB PENYEBARAN DANA BANTUAN OPERASIONAL SEKOLAH OPTICAL CHARACTER RECOGNITION MENGGUNAKAN ALGORITMA TEMPLATE MATCHING CORRELATION More cited articles

Dehghanian A, Ghods V. Farsi handwriting digit recognition based on convolutional neural networks. In2018 6th International Symposium on Computational and Business Intelligence (ISCBI) 2018 Aug 27 (pp. 65-68). IEEE. 10.1109/ISCBI.2018.00022
Zhang C, Zhou Z, Lin L. Handwritten digit recognition based on convolutional neural network. In2020 Chinese automation congress (CAC) 2020 Nov 6 (pp. 7384-7388). IEEE. 10.1109/CAC51589.2020.9326781
Ahlawat S, Rishi R. Handwritten digit recognition using adaptive neuro-fuzzy system and ranked features. In2018 International Conference on Computing, Power and Communication Technologies (GUCON) 2018 Sep 28 (pp. 1128-1132). IEEE. 10.1109/GUCON.2018.8675013
Kayumov Z, Tumakov D. Convolution neural network learning features for handwritten digit recognition. In2020 IEEE East-West Design & Test Symposium (EWDTS) 2020 Sep 4 (pp. 1-5). IEEE.10.1109/EWDTS50664.2020.9224822
M. Rajalakshmi, P. Saranya, and P. Shanmugavadivu, “Pattern Recognition-Recognition of Handwritten Document Using Convolutional Neural Networks,” IEEE Int. Conf. Intell. Tech. Control. Optim. Signal Process. INCOS 2019, pp. 1–7, 2019, doi: 10.1109/INCOS45849.2019.8951342.10.1109/INCOS45849.2019.8951342
Wisayataksin S, Boonyuu G. A programmable artificial neural network coprocessor for handwritten digit recognition. 2019 Int. Conf. Inf. Commun. Technol. ICOIACT 2019. 2019. p. 139-42. doi: 10.1109/ICOIACT46704.2019.8938541.10.1109/ICOIACT46704.2019.8938541
Ozyildirim BM. Handwritten Digits Classification with Generalized Classifier Neural Network. 2018 Innov. Intell. Syst. Appl. Conf. 2018. p. 1-3. 10.1109/ASYU.2018.8553999
Boukrouh M, Lashab M, Goutas A, Ouchtati S. Recognition of Handwritten Digits Using Neuro-Markovian Technique. p. 1-5. 10.1109/SIVA.2018.8661148
Shopon M, Mohammed N, Abedin MA. Image augmentation by blocky artifact in Deep Convolutional Neural Network for handwritten digit recognition. 2017 IEEE Int. Conf. Imaging, Vis. Pattern Recognition, icIVPR 2017. 2017. doi: 10.1109/ICIVPR.2017.7890867.10.1109/ICIVPR.2017.7890867
Sethi R, Kaushik I. Hand written digit recognition using machine learning. Proc. - 2020 IEEE 9th Int. Conf. Commun. Syst. Netw. Technol. CSNT 2020. 2020. p. 49-54. doi: 10.1109/CSNT48778.2020.9115746.10.1109/CSNT48778.2020.9115746
Choudhury A, Rana HS, Bhowmik T. Handwritten Bengali Numeral Recognition using HOG Based Feature Extraction Algorithm. 2018 5th Int. Conf. Signal Process. Integr. Networks, SPIN 2018. 2018. p. 687-90. doi: 10.1109/SPIN.2018.8474215.10.1109/SPIN.2018.8474215
Nyide O, Gwetu MV. Handwritten digit classification using iterative haar-like operators and neural networks. icABCD 2019 - 2nd Int. Conf. Adv. Big Data, Comput. Data Commun. Syst. 2019. p. 1-5. doi: 10.1109/ICABCD.2019.8850998.10.1109/ICABCD.2019.8850998
Islam KT, Mujtaba G, Raj RG, Nweke HF. Handwritten digits recognition with artificial neural network. 2017 Int. Conf. Eng. Technol. Technopreneurship, ICE2T 2017. 2017. p. 1-4. doi: 10.1109/ICE2T.2017.8215993.10.1109/ICE2T.2017.8215993
Nandan KVP, Panda M, Veni S. Handwritten digit recognition using ensemble learning. Proc. 5th Int. Conf. Commun. Electron. Syst. ICCES 2020. 2020. p. 1008-13. doi: 10.1109/ICCES48766.2020.09137933.10.1109/ICCES48766.2020.9137933
Shrivastava A, Jaggi I, Gupta S, Gupta D. Handwritten Digit Recognition Using Machine Learning: A Review. 2019 2nd Int. Conf. Power Energy Environ. Intell. Control. PEEIC 2019. 2019. p. 322-6. doi: 10.1109/PEEIC47157.2019.8976601.10.1109/PEEIC47157.2019.8976601
Mehta M, Bhadauria M. Performance Evaluation of Neural Network for Human Classification Using Blob Dataset. Recent Advances in Computer Science and Communications. 2019;13. doi: 10.2174/2666255813666190912101628.https://doi.org/10.2174/2666255813666190912101628
Mehta M, Bhadauria M, Verma K. Detecting Human from Environment: Using AI for embellishing preprocessing step of Human Accedence. 2015
Mehta M, Kaur K. An Automated Approach to Detect and Label Abandoned Objects from Videos using Generalized ROIs. 2020 IEEE 17th India Council International Conference (INDICON). 2020. p. 1-6. doi: 10.1109/INDICON49873.2020.9342599
Mehta M, Bhadauria M. A perlustration of Human Apprehension and Behavior Accedence in surroundings. 2014;8:38-43
Mehta, Monika, and Madhulika Bhatia. "Extraction of posture silhouettes using human posture feature points and spatial masks for activity recognition in public place." Journal of Engineering Research 10.3A (2022). https://doi.org/10.36909/jer.10803
Monika, and Kamaldeep Kaur. "Reproducibility of AOD Algorithm: An Experimental evaluation for Key-Predictors Identification." EAI Endorsed Trans. Context aware Syst. Appl. 7.20 (2020): e3. DOI: 10.4108/eai.13-7-2018.164099
M. N. Baihaqi, A. Sugiharto, and H. Tantyoko, "Classification of Real and Fake Images Using Error Level Analysis Technique and MobileNetV2 Architecture," Jurnal Masyarakat Informatika, vol. 16, no. 1, pp. 54 - 68, May. 2025. https://doi.org/10.14710/jmasif.16.1.73283
M. Pradnyatama, C. A. Sari, E. H. Rachmawanto, and H. M. M. Islam, "A Comparative Analysis of Convolutional Neural Network (CNN): MobileNetV2 and Xception for Butterfly Species Classification," Jurnal Masyarakat Informatika, vol. 16, no. 1, pp. 69 - 90, May. 2025. https://doi.org/10.14710/jmasif.16.1.72957

Last update:

No citation recorded.

Last update: 2026-05-06 14:37:20

No citation recorded.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

The authors who submit the manuscript must understand that the article's copyright belongs to the author(s) if accepted for publication. However, the author(s) grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. Authors should also understand that their article (and any additional files, including data sets, and analysis/computation data) will become publicly available once published under that license. By submitting the manuscript to Jmasif, the author(s) agree with this policy. No special document approval is required.

The author(s) guarantee that:

their article is original, written by the mentioned author(s),
has never been published before,
does not contain statements that violate the law, and
does not violate the rights of others, is subject to copyright held exclusively by the author(s), is free from the rights of third parties, and the necessary written permission to quote from other sources has been obtained by the author(s).

The author(s) retain all rights to the published work, such as (but not limited to) the following rights:

Copyright and other proprietary rights related to the article, such as patents,
The right to use the substance of the article in its own future works, including lectures and books,
The right to reproduce the article for its own purposes,
The right to archive all versions of the article in any repository, and
The right to enter into separate additional contractual arrangements for the non-exclusive distribution of published versions of the article (for example, posting them to institutional repositories or publishing them in a book), acknowledging its initial publication in this journal (Jurnal Masyarakat Informatika).

Suppose the article was prepared jointly by more than one author. Each author submitting the manuscript warrants that all co-authors have given their permission to agree to copyright and license notices (agreements) on their behalf and notify co-authors of the terms of this policy. Jmasif will not be held responsible for anything arising because of the writer's internal dispute. Jmasif will only communicate with correspondence authors.

Authors should also understand that their articles (and any additional files, including data sets and analysis/computation data) will become publicly available once published. The license of published articles (and additional data) will be governed by a Creative Commons Attribution-ShareAlike 4.0 International License. Jmasif allows users to copy, distribute, display and perform work under license. Users need to attribute the author(s) and Jmasif to distribute works in journals and other publication media. Unless otherwise stated, the author(s) is a public entity as soon as the article is published.