Clustering for Categorical Data Using Nonlinear Goal Programming

نوع المستند : المقالة الأصلية

المؤلفون

1 Faculty of commerce, Al-Azhar University Girl's Branch, Cairo, Egypt.

2 Faculty of Economics and Political Science, Cairo University, Egypt.

10.21608/jsfc.2025.359333.1016

المستخلص

Clustering is a popular unsupervised learning method used to group similar data points together. However, handling categorical variables in clustering can be challenging, as most clustering algorithms are designed to work with numerical data. Mathematical programming is a systematic model used for minimizing or maximizing the value of an objective function with respect to a set of constraints. The study suggests a mathematical nonlinear goal programming model for qualitative data clustering. The study clustering qualitative data using the suggested nonlinear goal programming model which has three main advantages: first is the data that is directly used, without the need of being converted to quantitative values, second is the optimal clusters which are automatically obtained by solving the optimization problem and third cluster analysis as an exploratory tool to support the identification of associations within qualitative data, cluster analysis can be helpful in identifying patterns where numerous cases are studied. The study evaluates the performance of the suggested mathematical nonlinear goal programming model using numerical examples by preliminary cases. The results for the suggested mathematical nonlinear goal programming model has to be proved efficient with a general average purity 64.2%.

الكلمات الرئيسية

ملفات

شارك

إرسال الاستشهاد إلى

الإحصائيات