Grid-Based Classifier as a Replacement for Multiclass Classifier in a Supervised Non-Parametric Approach

Pattern recognition/classification has received a considerable attention in engineering fields. In most applications, it is desirable to maintain the classification accuracy, but also reduce the classification time. The quality of a given classification technique is measured by the computational complexity, execution time of algorithms, and the number of patterns that can be classified correctly despite any distribution. In this thesis, a new method known as Grid Based Classifier was proposed. This method carries the advantages of the two previous methods in order to improve the classification tasks. The problem with the current lazy algorithms is that they learn quickly, but classify very slowly. On the other hand, the eager algorithms classify quickly, but they learn very slowly. The two algorithms were compared, and the proposed algorithm was found to be able to both learn and classify quickly. The method was developed based on the grid structure, whereby it was done to create a successful method of improving performance in classification. In the current research, the new algorithm was tested and applied to the multiclass classification of two or more categories, which are important for handling problems related to practical classification. The new method was also compared with the Levenberg-Marquardt back-propagation neural network in the learning stage and the Condensed nearest neighbor in the testing stage to examine the performance of the model. The experimental results on artificial data sets and real-world data sets (from UCI Repository) show that the new method could improve both the efficiency and accuracy of pattern classification. In real-world experiment (Haberman data set), new method allows 1% improvement in training accuracy and 1.8% improvement in testing accuracy and also allows considerable improvement in running time comparing to neural network method.

Preview PDF ABS__FK_2009_83.pdf Download (618kB)

Actions (login required)

View Item