A novel hybrid partial least squares quadratic discriminant classification algorithm

Kyaw Kyaw Htike, “A novel hybrid partial least squares quadratic discriminant classification algorithm”, Contemporary Engineering Sciences, Vol. 9, no. 32, pp. 1547-1557, 2016. DOI: 10.12988/ces.2016.69153. [Scopus-indexed journal]

Abstract:

Giving computers and machines the ability to recognize and classify patterns is an important task in engineering systems with numerous applications in many aspects of life. The field of pattern classification has become very popular in the past decade and although numerous techniques have been proposed, there is still much room for improvement. In this paper, we propose a novel pattern classification algorithm which we term “Hybrid Partial Least Squares Quadratic Discriminant Classifier” that outperforms several state-of-the-art classifiers. Evaluating on three publicly available challenging datasets demonstrates the effectiveness of our proposed approach.

Keywords: classification, pattern recognition, hybrid, machine learning