Enhancing handwritten numeric string recognition through incremental support vector machines

Rani Kurnia Putri; Muhammad Athoillah

doi:10.59400/jam.v2i1.373

Enhancing handwritten numeric string recognition through incremental support vector machines

Rani Kurnia Putri Department of Mathematics Education, Universitas PGRI Adi Buana Surabaya, Jawa Timur 60234, Indonesia
Muhammad Athoillah Department of Statistics, Universitas PGRI Adi Buana Surabaya, Jawa Timur 60234, Indonesia http://orcid.org/0000-0002-1800-1314

Ariticle ID: 373

163 Views, 172 PDF Downloads

DOI: https://doi.org/10.59400/jam.v2i1.373

Keywords: digital image processing; incremental learning; pattern recognition; dynamic machine learning

Abstract

Handwritten digit recognition systems are integral to diverse applications such as postal services, banking, and document processing in our digitally-driven society. This research addresses the challenges posed by evolving datasets and dynamic scenarios in handwritten digit recognition by proposing an approach based on incremental support vector machines (ISVM). ISVM is an extension of traditional support vector machines (SVM) designed to handle scenarios where new data points become available over time. The dataset includes handwritten images (numbers “0” to “6”) and trials introducing new classes (“7”, “8”, and “9”). Evaluation utilizes k-fold cross-validation for robustness. Digital image processing involves converting images into numeric data using the histogram method. The result showed the positive outcomes of using ISVM in handwritten digit recognition, emphasizing its adaptability to incremental learning and its ability to maintain robust performance in the face of evolving datasets, which is crucial for real-world applications.

References

[1] Athoillah M, Putri RK. Handwritten Arabic Numeral Character Recognition Using Multi Kernel Support Vector Machine. Kinetik: Game Technology, Information System, Computer Network, Computing, Electronics, and Control. 2019; 4(2): 99-106. doi: 10.22219/kinetik.v4i2.724

[2] Memon J, Sami M, Khan RA, et al. Handwritten Optical Character Recognition (OCR): A Comprehensive Systematic Literature Review (SLR). IEEE Access. 2020, 8: 142642-142668. doi: 10.1109/access.2020.3012542

[3] Diaz M, Ferrer MA, Impedovo D, et al. A Perspective Analysis of Handwritten Signature Technology. ACM Computing Surveys. 2019, 51(6): 1-39. doi: 10.1145/3274658

[4] Athoillah M. K-Nearest Neighbor for Recognize Handwritten Arabic Character. Jurnal Matematika “MANTIK.” 2019, 5(2): 83-89. doi: 10.15642/mantik.2019.5.2.83-89

[5] Yogesh Y, Ghantasala GSP, Priya A. Artificial Intelligence Based Handwriting Digit Recognition (HDR) - A Technical Review. In: Proceedings of the 2023 International Conference on Device Intelligence, Computing and Communication Technologies, (DICCT); 17–18 March 2023; Dehradun, India. pp. 275–278. doi: 10.1109/dicct56244.2023.10110186

[6] Khobragade RN, Koli NA, Lanjewar VT. Challenges in recognition of online and off-line compound handwritten characters: A review. In: Zhang YD, Mandal J, So-In C, Thakur N (editors). Smart Trends in Computing and Communications. Springer; 2020. Volume 165. pp. 375–383. doi: 10.1007/978-981-15-0077-0_38

[7] Liu X, Lian Y. Handwriting identification: Challenges and solutions. Journal of Forensic Science and Medicine. 2018; 4(3): 167-173. doi: 10.4103/jfsm.jfsm_81_17

[8] Yang Q, Gu Y, Wu D. Survey of incremental learning. In: Proceedings of the 2019 Chinese Control And Decision Conference (CCDC); 3–5 June 2019; Nanchang, China. pp. 399–404. doi: 10.1109/ccdc.2019.8832774

[9] Nayak J, Naik B, Behera HS. A Comprehensive Survey on Support Vector Machine in Data Mining Tasks: Applications & Challenges. International Journal of Database Theory and Application. 2015, 8(1): 169-186. doi: 10.14257/ijdta.2015.8.1.18

[10] Cervantes J, Garcia-Lamont F, Rodríguez-Mazahua L, et al. A comprehensive survey on support vector machine classification: Applications, challenges and trends. Neurocomputing. 2020, 408: 189-215. doi: 10.1016/j.neucom.2019.10.118

[11] Shams M, Amira A, Wael Z. Arabic Handwritten Character Recognition based on Convolution Neural Networks and Support Vector Machine. International Journal of Advanced Computer Science and Applications. 2020, 11(8). doi: 10.14569/ijacsa.2020.0110819

[12] Rajnoha M, Burget R, Dutta MK. Offline handwritten text recognition using support vector machines. In: Proceedings of the 2017 4th International Conference on Signal Processing and Integrated Networks (SPIN); 2–3 February 2017; Noida, India. pp. 132–136. doi: 10.1109/spin.2017.8049930

[13] Katiyar G. Off-Line Handwritten Character Recognition System Using Support Vector Machine. American Journal of Neural Networks and Applications. 2017, 3(2): 22. doi: 10.11648/j.ajnna.20170302.12

[14] Athoillah M, Imah EM, Irawan MI. Multi kernel-based classification method with incremental learning for image retrieval. In: Proceedings of the International Conference on Mathematics: Pure, Applied and Computation, ICoMPAC 2015; 3 December 2015; Surabaya, Indonesia.

[15] Athoillah M. Face Recognition Using Multi Kernel SVM with Incremental Learning (Malaysian). Jurnal Online Informatika. 2018, 2(2): 84. doi: 10.15575/join.v2i2.109

[16] Peryanto A, Yudhana A, Umar R. Image Classification Using Convolutional Neural Network and K Fold Cross Validation (Indonesian). Journal of Applied Informatics and Computing. 2020, 4(1): 45-51. doi: 10.30871/jaic.v4i1.2017

[17] Xu Y, Goodacre R. On Splitting Training and Validation Set: A Comparative Study of Cross-Validation, Bootstrap and Systematic Sampling for Estimating the Generalization Performance of Supervised Learning. Journal of Analysis and Testing. 2018, 2(3): 249-262. doi: 10.1007/s41664-018-0068-2

[18] Zhang Y, Yang Y. Cross-validation for selecting a model selection procedure. Journal of Econometrics. 2015, 187(1): 95-112. doi: 10.1016/j.jeconom.2015.02.006

[19] Braga-Neto U. Fundamentals of Pattern Recognition and Machine Learning. Springer International Publishing, 2020. doi: 10.1007/978-3-030-27656-0

[20] Ye Q, Doermann D. Text Detection and Recognition in Imagery: A Survey. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2015, 37(7): 1480-1500. doi: 10.1109/tpami.2014.2366765

[21] Ajam A, Forghani M, AlyanNezhadi MM, et al. Content-based image retrieval using color difference histogram in image textures. In: Proceedings of the 2019 5th Iranian Conference on Signal Processing and Intelligent Systems (ICSPIS); 18–19 December 2019; Shahrood, Iran. pp. 1–6. doi: 10.1109/icspis48872.2019.9066062

[22] Somvanshi M, Chavan P, Tambade S, et al. A review of machine learning techniques using decision tree and support vector machine. In: Proceedings of the 2016 International Conference on Computing Communication Control and automation (ICCUBEA); 12–13 August 2016; Pune, India. pp. 1–7. doi: 10.1109/iccubea.2016.7860040

[23] Chefrour A. Incremental supervised learning: algorithms and applications in pattern recognition. Evolutionary Intelligence. 2019, 12(2): 97-112. doi: 10.1007/s12065-019-00203-y

[24] Pisner DA, Schnyer DM. Support vector machine. Machine Learning: Methods and Applications to Brain Disorders. Academic Press; 2020. pp. 101–121. doi: 10.1016/b978-0-12-815739-8.00006-7

[25] Campbell C, Ying Y. Learning with Support Vector Machines. Springer Nature; 2022.

[26] Zhang YC, Hu GS, Zhu FF, et al. A new incremental learning support vector machine. In: Proceedings of the 2009 International Conference on Artificial Intelligence and Computational Intelligence; 7–8 November 2009; NW Washington, DC, United States. doi: 10.1109/aici.2009.342

[27] Juba B, Le HS. Precision-Recall versus Accuracy and the Role of Large Data Sets. Proceedings of the AAAI Conference on Artificial Intelligence. 2019, 33(01): 4039-4048. doi: 10.1609/aaai.v33i01.33014039

Published

2024-01-19

How to Cite

Putri, R. K., & Athoillah, M. (2024). Enhancing handwritten numeric string recognition through incremental support vector machines. Journal of AppliedMath, 2(1), 373. https://doi.org/10.59400/jam.v2i1.373

Download Citation

Issue

Vol. 2 No. 1 (2024)

Section

Article

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

Editor-in-Chief

Assoc. Prof. Ahmad Zeeshan

International Islamic University, Pakistan

eISSN

2972-4805

Publication Frequency

Bi-monthly

About the Publisher

Academic Publishing insists on taking academic exchange and publication as the main line, carrying out comprehensive management based on science and technology, and fully exploring excellent international publishing resources. Within 5 years, it will form a strategic framework and scale with science (S), technology (T), medicine (M), education (E), and humanities and arts (H) as the main publishing fields. Academic Publishing is headquartered in Singapore and based in Malaysia, with the United States and China providing the main scientific and academic resources. At the same time, it has established long-term good cooperative relations with other publishing companies, scientific research communities, and academic organizations in more than a dozen countries and regions. Academic Publishing uses English and Chinese as its main publishing languages, mainly publishing books, journals, and conference papers in print and online. The vast majority of publications follow the international open access policy, providing stable and long-term quality and professional publications. With the joint efforts of the expert team and our professional editorial team, our publications will gradually be indexed by international databases in stages to provide convenient and professional retrieval for various scholars. At the same time, manuscripts we accept will be subject to the peer review principle, and cutting-edge and innovative research articles will be preferentially accepted for peer reference and discussion. All kinds of our publications are welcome for peer to contribute, access, and download.

more

Volume Arrangement

2024

2023

Featured Articles

Octagonal-square tessellation model for masting GSM network: A case study of MTN Kumasi-East, Ghana

Masting in GSM network design is one of the most challenging problems in cell planning. The effect of uniform design pattern has been proven geographically to be hexagonal using uniform cell range. In this paper, we present a new uniform greedy semi-regular tessellation model called the octagonal square tessellation model (OSTM) to address the problem of global minimum overlap difference and area. Data from MTN Kumasi-East Ghana was collected and analyzed using the developed model. The original layout for the 0.6 km cell range accounted for an overlap difference of 937.66 m and a total area of 21.41 km² for 50 GSM mosts whereas the OSTM model accounted for an overlap difference of 1316.95 m with an area of 34.23 km². This is a 59.87% reduction of the original total area. Our solution is shown to be optimal in overlap difference and area for non-uniform cell range.

Computation of topological indices of linear chains of perylene and coronene using M-polynomial

The M-polynomial yields degree based topological indices that anticipate different physical and chemical properties of material being scrutinized. In this work, M- polynomial of linear chains of perylene and coronene are acquired. From M-polynomial, some degree based topological dicriptors are determined. Some topological indices of these compounds are compared by plotting graphs.

H∞ hybrid control and MRD in a steel frame building subjected to excessivevibrations caused by the dynamic action of wind and earthquake

The dynamic loads from earthquakes and winds can destroy lives, cause collapse in civil structures, and interrupt basic services provided to the population. In this scenario, structural designs must be developed to decrease the damage induced by these actions. The objective of this work is to design a hybrid controller based on the H∞ optimization via state feedback and the magneto-rheological damper (MRD) to mitigate the excessive vibrations of a three-story steel frame building, represented through the shear building model, subjected to the simultaneous dynamic action of wind and earthquake. All research is based on computational simulation, experimental research and results will not be addressed. In the numerical analysis, digital computer and MATLAB® software are used, and implemented codes generate the expected results based on the mathematical modeling. With the application of the H∞ control technique via state feedback, the displacements were reduced by 77%. With MRD this reduction was 79%. With the hybrid controller, this reduction was 100%. Thus, the verifications in relation to maximum displacements were met for NBR 15421:2006, NBR 8800:2008 and NBR 6118:2014. From the results, it is concluded that the hybrid controller proved to be more efficient and achieved the proposed objective. The exogenous inputs had zero influence on the behavior of the system output.

Topological analysis of multiple tables

The paper proposes a topological approach in order to explore several data tables simultaneously. These data tables of quantitative and/or qualitative variables measured on different homogeneous themes, collected from the same individuals. This approach, called topological analysis of multiple tables (TAMT), is based on the notion of neighborhood graphs in the context of a joint analysis of several data tables. It allows the simultaneous study of possible links between several thematic tables. The structure of the correlations or associations of the variables in each thematic table is analyzed according to quantitative, qualitative or mixed variables considered. Like the multiple factorial analysis (MFA), the TAMT allows several tables of variables to be analyzed simultaneously, and to obtain results, in particular graphical representations, which make it possible to study the relationship between individuals, variables and tables of data. These can also be tables of temporal data, collected at different times on the same individuals. The proposed TAMT approach is illustrated using real data associated with several and different homogeneous themes. Its results are compared to those from the MFA method.

Development of a stacked hybrid Decision Tree model leveraging the NSL-KDD dataset

Intrusion detection in information technology as well as operational technology networks is highly required in modern day systems due to the increased spate of cyber-attacks in both number and complexity. Anomaly-based intrusion detection systems which have the capacity to detect novel or zero-day attacks are highly employed in this regard. One important component of anomaly-based intrusion detection systems which ensures their behaviour is artificial intelligence in general and machine learning in particular. The burden in modern day cybersecurity research is to investigate and develop models that can outperform existing ones. This paper is aimed at developing a hybrid decision tree model using the stacking ensemble approach. Performances were measured on the basis of recall, precision, accuracy, F1-score, receiver operating characteristics and confusion matrices. The hybrid model presented a precision of 97%, accuracy of 81%, F1-score of 80% and AUC score of 0.96, respectively.