Paper Title
Classifier Combination for Arabic Character Recognition

Abstract
Arabic character recognition continues to be among the most challenging character recognition systems. A traditional Arabic character recognition problem is solved in five steps. This paper is interested in the last step called classification or recognition. We propose a classification strategy based on the parallel combination of classifiers. It is called Borda count. Borda count is implemented using the artificial neural network and the K-nearest neighbors. This method significantly improves the recognition rate compared to the recognition rate delivered by a single classifier. Keywords - Artificial Intelligence, Machine Learning, Classification, Character Recognition.