Data analysis and prediction model discovery through decision tree learning. (30th May 2023)

• Record Type: Journal Article
• Authors: Porouhan, Parham
• Is Part Of: International journal of experimental design and process optimisation. Volume 7:Number 1(2022)
• Page Start: 18
• Page End: 30
• Publication Date: 2023-05-30
• Subjects: data mining -- rapidminer studio -- decision trees -- predictive analytics -- model discovery -- medical dataset; Experimental design -- Periodicals; Process control -- Statistical methods -- Periodicals; 620.0072
• Languages: English
• ISSNs: 2040-2252

Abstract

Data mining and machine learning tools and approaches have been extensively used and practiced in bioinformatics with the purpose of analysing biomedical datasets. In this paper, we have chosen 'RapidMiner Studio' as a platform to create/generate 'decision tree models', which are obtained from a medical dataset. In research, 'decision trees' are considered as one of the most popular supervised (machine) learning techniques, which, as an instance can be effectively used for prediction of specific indicators/attributes affecting a disease. In general, this paper is divided into two main parts. In the first part of the study, an existing medical dataset with known label values (so-called, a training dataset) is used to identify and discover the hidden patterns. In the second part of the study, the generated/resulting patterns are used for making predictions. The results of the study provide groundwork for further and future studies.