Akademik Veri Yönetim Sistemi
7th International Researchers, Statisticians, and Young Statisticians Congress 2nd – 5th November 2023, İstanbul, Türkiye, 2 Kasım - 05 Aralık 2023, ss.165-166
The Nearest Neighbour method has been widely used in machine learning and data mining
because of how easy it is to use and how well it categorizes independent observations. Its
inability to take into consideration interdependencies between observations, however, has
restricted its application to dependent data structures. In this study, a generalized structure for
the Nearest Neighbour method is presented that is specifically designed for dependent
observations. We supply an optimized method that improves classification accuracy by
utilizing a brand-new metric that takes into account both the interdependencies between
observations and the individual features. It has been proven through comprehensive research,
including cross-validation and performance metrics, that our approach performs better when
managing dependent data sets than conventional Nearest Neighbour methods. By providing a
reliable method for data sets where connections between observations cannot be ignored, our
research closes a sizable gap in the literature.
.
Keywords: Nearest Neighbour Approach; Dependent Observations; Generalized Framework; Data
Interdependencies; Classification Accuracy.