1. MMC
  2. All Volumes and Issues
  3. Volume 12, Number 2, 2025
  4. Comparing persistent homology-based classifiers for Filipino sign language recognition

Comparing persistent homology-based classifiers for Filipino sign language recognition

MMC.
2025;
Volume 12, Number 2
: pp. 512–524
https://doi.org/10.23939/mmc2025.02.512
Received: December 26, 2024
Revised: May 24, 2025
Accepted: May 25, 2025

Jetomo C. B., De Lara M. L. D.  Comparing persistent homology-based classifiers for Filipino sign language recognition.  Mathematical Modeling and Computing. Vol. 12, No. 2, pp. 512–524 (2025)  

Authors:
  1. C. B. Jetomo
  2. M. L. D. De Lara
1
Institute of Mathematical Sciences, University of the Philippines Los Banos
2
Institute of Mathematical Sciences, University of the Philippines Los Banos

Deaf or hard of hearing individuals have long been faced with problems in communication.  To cope with this communication gap, numerous sign languages have been developed, one of which is the Filipino Sign Language (FSL). Despite FSL being declared as the national sign language of the Philippines, there is lack of formal implementation of policies and the gap problem continues to prevail.  Sign interpreters play a crucial role in this limitation but are still insufficient in number.  Hence, machine learning techniques are leveraged to automate the interpretation process of signed gestures and the field of Sign Language Recognition (SLR) is developed.  This paper extends this by utilizing computational topology-based methods in performing SLR on a FSL dataset.  Specifically, it aims to utilize the Persistent Homology Classification Algorithm (PHCA) in classifying or interpreting dynamic FSL gestures.  Due to many distinct classes considered in this problem, this paper also aims to develop the multi-level PHCA in which this approach divides the classification task.  It does so by defining categories consisting of classes and performing two-part classification, first on the categories, and second on the classes within the selected category.  The performance of PHCA and multi-level PHCA using different classification metrics is evaluated.  The predicting and training time of the models are also compared. Results show that both PHCA and multi-level PHCA produced satisfactory performance for SLR.  It is shown further that multi-level PHCA outperformed PHCA in all setups considered, producing an accuracy of 85% for 10 classes and 60.99% on 105 classes, indicating a potential for further research.

Filipino sign language
persistent homology
sign language recognition
multi-level classification
  1. Davis A. C., Hoffman H. J.  Hearing loss: rising prevalence and impact.  Bulletin of the World Health Organ.  97 (10), 646–646A (2019).
  2. Most T.  The effects of degree and type of hearing loss on children's performance in class.  Deafness & Education International.  6 (3), 154–166 (2004).
  3. Cruz F. W. S.-d., Calimpusan E. C.  Status and challenges of the deaf in one city in the Philippines: towards the development of support systems and socio-economic opportunities.  Asia Pacific Journal of Multidisciplinary Research.  6 (2), 61–74 (2018).
  4. River J. P., Ong C.  Recognizing non-manual signals in Filipino sign language.  Proceedings of the LREC2018 8th Workshop on the Representation and Processing of Sign Languages: Involving the Language Community.  177–184 (2018).
  5. Cabalfin E. P., Martinez L. B., Guevara R. C. L., Naval P. C.  Filipino sign language recognition using manifold projection learning.  TENCON 2012 IEEE Region 10 Conference.  1–5 (2012).
  6. Montefalcon M. D., Padilla J., Rodriguez R.  Filipino Sign Language Recognition Using Long Short-Term Memory and Residual Network Architecture.  Proceedings of Seventh International Congress on Information and Communication Technology.  489–497 (2023).
  7. Montefalcon M. D., Padilla J. R., Llabanes R. R.  Filipino Sign Language Recognition Using Deep Learning.  2021 5th International Conference on E-Society, E-Education and E-Technology.  219–225 (2021).
  8. Sandjaja I., Marcos N.  Sign Language Number Recognition.  2009 Fifth International Joint Conference on INC, IMS and IDC.  1503–1508 (2009).
  9. Oliva K. E., Ortaliz L. L, Tobias M. A., Vea L.  Filipino Sign Language Recognition for Beginners using Kinect.  2018 IEEE 10th International Conference on Humanoid, Nanotechnology, Information Technology,Communication and Control, Environment and Management (HNICEM).  1–6 (2018).
  10. Hensel F., Moor M., Rieck B.  A Survey of Topological Machine Learning Methods.  Frontiers in Artificial Intelligence.  4, 1–12 (2021).
  11. Dirafzoon A., Lokare N., Lobaton E.  Action classification from motion capture data using topological data analysis.  2016 IEEE Global Conference on Signal and Information Processing (GlobalSIP).  1260–1264 (2016).
  12. Yang H., Sun D., Cai Y.-J., Yang J., Si X.-Y., Zhou S.-M., Yan Y.  Learning Topological Representation of 3D Skeleton Dynamics with Persistent Homology for Human Activity Recognition.  2023 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).  2709–2716 (2023).
  13. Yan S., Xiong Y., Lin D.  Spatial temporal graph convolutional networks for skeleton-based action recognition.  Proceedings of the AAAI Conference on Artificial Intelligence.  32 (1), 7444–7452 (2018).
  14. De Lara M. L. D.  Persistent homology classification algorithm.  PeerJ Computer Science.  9, e1195 (2023).
  15. Tupal I. J.  FSL-105: A dataset for recognizing 105 Filipino sign language videos.  Mendeley Data, V2 (2023).
  16. Caliwag A. C., Hwang H.-J., Kim S.-H., Lim W.  Movement-in-a-Video Detection Scheme for Sign Language Gesture Recognition Using Neural Network.  Applied Sciences.  12 (20), 10542 (2022).
  17. Kwon O., Sim J. M.  Effects of data set features on the performances of classification algorithms.  Expert Systems with Applications.  40 (5), 1847–1857 (2013).
  18. Tupal I., Cabatuan M., Manguerra M.  Recognizing Filipino Sign Language with InceptionV3, LSTM, and GRU.  2022 IEEE 14th International Conference on Humanoid, Nanotechnology, Information Technology, Communication and Control, Environment, and Management (HNICEM).  1–5 (2022).
  19. Carlsson G.  Topology and data.  Bulletin of the American Mathematical Society.  46, 255–308 (2009).
  20. Edelsbrunner H.  Computational Topology: An Introduction.  AMS  (2008).
  21. Eldesbrunner H., Harer J.  Persistent homology – a survey.  Surveys on Discrete and Computational Geometry: Twenty Years Later. Contemporary Mathematics.  257–282 (2008).
  22. Moral P. D., Nowaczyk S., Pashami S.  Why is Multiclass Classification Hard?  IEEE Access.  10, 80488–80462 (2022).
  23. Hameed N., Shabut A. M., Ghosh M. K., Hossain M. A.  Multi-class multi-level classification algorithm for skin lesions classification using machine learning techniques.  Expert Systems with Applications.  141, 112961 (2020).
  24. Zhang X., Liu C.-A.  Model averaging prediction by K-fold cross-validation.  Journal of Econometrics.  235 (1), 280–301 (2023).
  25. Ahsan M. M., Mahmud M. A. P., Saha P. K., Gupta K. D., Siddique Z.  Effect of data scaling methods on machine learning algorithms and model performance.  Technologies.  9 (3), 52 (2021).
  26. Jolliffe I. T., Cadima J.  Principal component analysis: a review and recent developments.  Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences.  374 (2065), 20150202 (2016).