Machine Learning Techniques for Handwritten Digit Recognition

The following study examines how various classification algorithms perform on the problem of handwritten digit recognition. The classifiers discussed are k-Nearest Neighbour (k-NN), Single Classification Decision Trees and Bagged Decision Trees. These algorithms were evaluated with the use of information from the United States Postal Service (USPS). This study’s results show that the k-NN classifier had the fastest performance while the bagged decision trees were the slowest. In terms of classification performance, the bagged decision tree method was found to have the fewest misclassifications and outperformed k-NN and single classification trees in all of the considered metrics.

[1]  Kavita Mittal,et al.  Performance study of K-nearest neighbor classifier and K-means clustering for predicting the diagnostic accuracy , 2018, International Journal of Information Technology.

[2]  Ahmad Taher Azar,et al.  Covering-based rough set classification system , 2016, Neural Computing and Applications.

[3]  Krupa Dholakia,et al.  A Survey on Handwritten Character Recognition Techniques for Various Indian Languages , 2015 .

[4]  Zhiwen Yu,et al.  A survey on ensemble learning , 2019, Frontiers of Computer Science.

[5]  Aboul Ella Hassanien,et al.  Hybrid rough-bijective soft set classification system , 2016, Neural Computing and Applications.

[6]  Ahmad Taher Azar,et al.  Superior neuro-fuzzy classification systems , 2013, Neural Computing and Applications.

[7]  Ahmad Taher Azar,et al.  Rough set theory with Jaya optimization for acute lymphoblastic leukemia classification , 2019, Neural Computing and Applications.

[8]  Ahmad Taher Azar,et al.  Supervised hybrid feature selection based on PSO and rough sets for medical diagnosis , 2014, Comput. Methods Programs Biomed..

[9]  Ahmad Taher Azar,et al.  Tolerance rough set firefly-based quick reduct , 2016, Neural Computing and Applications.

[10]  Saso Dzeroski,et al.  MetaBags: Bagged Meta-Decision Trees for Regression , 2018, ECML/PKDD.

[11]  Ahmad Taher Azar,et al.  Hybrid Tolerance Rough Set: PSO Based Supervised Feature Selection for Digital Mammogram Images , 2013, Int. J. Fuzzy Syst. Appl..

[12]  Ahmad Taher Azar,et al.  Improved dominance rough set-based classification system , 2017, Neural Computing and Applications.

[14]  Quanmin Zhu,et al.  Complex System Modelling and Control Through Intelligent Soft Computations , 2016, Studies in Fuzziness and Soft Computing.

[15]  Ahmad Taher Azar,et al.  Fuzzy firefly clustering for tumour and cancer analysis , 2017 .

[16]  Sing H. Lee Optical Pattern Recognition , 1985, COMPCON.

[17]  K. Hornik,et al.  Unbiased Recursive Partitioning: A Conditional Inference Framework , 2006 .

[18]  Ching Y. Suen,et al.  Character Recognition Systems: A Guide for Students and Practitioners , 2007 .

[19]  Nafiz Arica,et al.  An overview of character recognition focused on off-line handwriting , 2001, IEEE Trans. Syst. Man Cybern. Syst..

[20]  Shereen M. El-Metwally,et al.  Decision tree classifiers for automated medical diagnosis , 2013, Neural Computing and Applications.

[21]  Soumya K. Ghosh,et al.  Optical Character Recognition Systems for Different Languages with Soft Computing , 2016, Studies in Fuzziness and Soft Computing.

[22]  Ahmad Taher Azar,et al.  Probabilistic neural network for breast cancer classification , 2012, Neural Computing and Applications.

[23]  Yun Li,et al.  Feature Selection for Identifying Critical Variables of Principal Components Based on K-Nearest Neighbor Rule , 2007, VISUAL.

[24]  Ethem Alpaydin,et al.  Bagging Soft Decision Trees , 2016, Machine Learning for Health Informatics.

[25]  Ahmad Taher Azar,et al.  Feature selection using swarm-based relative reduct technique for fetal heart rate , 2014, Neural Computing and Applications.

[26]  Ronald L. Rivest,et al.  Constructing Optimal Binary Decision Trees is NP-Complete , 1976, Inf. Process. Lett..

[27]  Rashmi Agrawal,et al.  Comparative Analysis of Decision Tree Algorithms , 2018 .