Samet Memiş - Department of Computer Engineering Faculty of Engineering and Natural Sciences İstanbul Rumeli University İstanbul, Turkey
Serdar Enginoğlu - Department of Mathematics Faculty of Arts and Sciences Çanakkale Onsekiz Mart University Çanakkale, Turkey
Uğur Erkan - Department of Computer Engineering Faculty of Engineering Karamanoğlu Mehmetbey University Karaman, Turkey

This paper introduces a new similarity measure of fuzzy parameterized fuzzy soft matrices (fpfs-matrices), i.e., Jaccard pseudo-similarity of fpfs-matrices. We then provide its basic properties. Afterwards, we apply it to the diagnosis of Parkinson’s Disease (PD), improving a machine learning (ML) approach. Next, we compare our approach with the well-known ML approaches, such as Naïve Bayes, 𝒌-Nearest Neighbor (𝒌NN), Support Vector Machine (SVM), Fuzzy 𝒌NN, Decision Trees (DT), Boosted Trees (BT), Adaptive Boosting Tree (AdaBoost), and Random Forest (RF) in terms of accuracy, specificity, and sensitivity. The results manifest that the proposed approach makes a more accuratediagnosis of PD than the others.

soft sets; fpfs-matrices; similarity measure;Parkinson’s disease; machine learning