Evaluation of Ensemble Classifiers for Handwriting Recognition

Abstract

-One of the major developments in machine learning in the past decade is the ensemble method, which finds highly accurate classifier by combining many moderately accurate component classifiers. In this research work, new ensemble classification methods are proposed for homogeneous ensemble classifiers using bagging and heterogeneous ensemble classifiers using arcing classifier and their performances are analyzed in terms of accuracy. A Classifier ensemble is designed using Radial Basis Function (RBF) and Support Vector Machine (SVM) as base classifiers. The feasibility and the benefits of the proposed approaches are demonstrated by the means of real and benchmark data sets of recognizing totally unconstrained handwritten numerals. The main originality of the proposed approach is based on three main parts: preprocessing phase, classification phase and combining phase. A wide range of comparative experiments are conducted for real and benchmark data sets of recognizing totally unconstrained handwritten numerals. The accuracy of base classifiers is compared with homogeneous and heterogeneous models for data mining problem. The proposed ensemble methods provide significant improvement of accuracy compared to individual classifiers and also heterogeneous models exhibit better results than homogeneous models for real and benchmark data sets of recognizing totally unconstrained handwritten numerals.

Providing secure and reliable communication for next generation networks in smart cities

“…A novel classifier-ensemble for intrusion detection by adopting Particle swarm optimization (PSO) weights has been proposed in [49]. Furthermore, the authors in [50] proposed a new hybrid detection system by adopting the Support Vector Machine (SVM) and Radial Basis Function (RBF). Their work proved the effectiveness of heterogeneous models in comparison to homogeneous solutions.…”

“…This chosen subset is the needed data to formulate the separating surface which as a result generates the support vector set. In non-linear SVM, the input vector is mapped into a high dimensional feature space, such that SVM generates a linear boundary in between various classes and maximizes the margin by adjusting the generated boundary [54]. It also maximizes the classification by subdividing the feature space into sub-spaces.…”

“…In our proposed binary classification technique, the data points are represented as shown in (11) [54]: = {( r , r ), ( m , m ), … , ( V , V )} (11) where ∈ } and ∈ {−1, +1}, refers to the binary value representing the classes, refers to the input vector, } refers to the whole dataset and is the number of training sets.…”

“…The objective is to figure out the hyper-plane that achieves the maximum margin. All training data falls under some restrictions as shown below [54].…”

Reliable Devanagri Handwritten Numeral Recognition using Multiple Classifier and Flexible Zoning Approach

“…Agnihotri et al [6] implemented a system for Devanagri handwritten recognition using diagonal features and genetic algorithm for a dataset of 1000 samples and obtained 85.78 % match score. Govindarajan [7] applied classifier Ensemble approach for handwritten Numeral recognition for NIST dataset. They used the combination of RBF and SVM and shown an improve result of 99.3% when the combination is used over the single classifier.…”