DISEASE CLASSIFICATION USING SUPPORT VECTOR MACHINE (SVM) WITH JAVA STANDARD EDITION (JSE)

Abstract

This research focuses on disease clustering, which is a crucial aspect of effective diagnosis and treatment. With the increasing complexity of health data generated from various sources, such as electronic health records and laboratory results, efficient methods are needed to cluster and analyze this data. The use of machine learning algorithms, particularly Support Vector Machine (SVM), offers a promising solution to address this issue. SVM is known for its ability to handle multidimensional data and identify patterns that are not immediately visible. The challenges faced in disease clustering include difficulties in managing large and complex data, as well as the inability of traditional methods to provide accurate and rapid results. Additionally, many healthcare professionals lack access to adequate analytical tools, hindering appropriate clinical decision-making. Therefore, it is essential to develop solutions that can effectively assist in disease clustering. The proposed solution in this study is the development of a Java Standard Edition (JSE) based application that implements the SVM algorithm for disease clustering. This application is designed to provide an intuitive user interface, allowing users to upload data, run the SVM algorithm, and easily obtain clustering results. This research uses clinical data from various diseases, including heart disease, diabetes, hypertension, cancer, asthma, and stroke. Evaluation results show that SVM can cluster diseases with an accuracy of up to 92%. Thus, this study concludes that the application of SVM in a JSE-based application is an effective solution for enhancing disease clustering and supporting better clinical decision-making.