Citazione: M. POPOLIZIO, A. AMATO, V. PIURI, V. DI LECCE: Improving Classification Performance Using the Semi-pivoted QR Approximation Algorithm. In: Rathore, V.S., Sharma, S.C., Ta vares, J.M.R., Moreira, C., Surendiran, B. (eds) Rising Threats in Expert Applications and Solutions. Lecture Notes in Networks and Systems, vol 434. Springer Nature Singapore, pp.263-271 2022 - https://link.springer.com/chapter/10.1007/978-981-19-1122-4_29
Abstract: Aim of this paper is to present a method to improve the classification performance of a Fuzzy C-means based classifier. The obtained results show that this method can improve the performance of the classifier both in terms of computational efficiency (by reducing the amount of data to be analyzed) and in terms of classification error rate. The proposed method is based on the Semi-Pivoted QR approximation (SPQR) algorithm. It reduces a numeric dataset (a matrix) to its more important features (where each feature is a column of the matrix). The framework discussed in this article can be used by researchers and practitioners to set up high-performance machine learning systems.
Keyword: Semi-pivoted QR approximation (SPQR) · Fuzzy C-means classifier · Data quality