Research Area:  Machine Learning
The main aim of this paper is to investigate various data mining and machine learning techniques employed for the analysis of rheumatoid arthritis prediction based on clinical and genetic factors. The clinical characters and gene factors are collected from various hospitals in Coimbatore region through laboratory investigations from the blood serum samples and general investigations. Patients with viral fever more than six weeks and later arthritis affected compared with those patients with viral fever and no rheumatoid arthritis developed. This study involves detailed analysis of machine learning algorithms employed for rheumatoid arthritis disease, and genetic factors involved in this disease. The relevant attributes taken from the literature and consultation of rheumatologists, a combination of clinical and genetic factors evolved in this disease. The proposed model works in a big data environment named Machine Learning based Ensemble Analytic Approach (MLEAA) consists of two phases, namely learning phase and prediction phase. In learning phase data’s are processed by map reduce framework in hadoop and the featured attributes are working towards prediction phase. The proposed MLEAA approach prediction phase consists of three different algorithms, namely Ababoost, SVM, ANN and based on voting system final predictive value is calculated. From this study achieve better results and it will be very useful for predict rheumatoid arthritis earlier.
Keywords:  
rheumatoid arthritis
data mining
machine learning
genetic factor
prediction phase
Author(s) Name:  S. Shanmugam & J. Preethi
Journal name:  Cognitive Science and Artificial Intelligence
Conferrence name:  
Publisher name:  Springer
DOI:  https://doi.org/10.1007/978-981-10-6698-6_7
Volume Information:  -
Paper Link:   https://link.springer.com/chapter/10.1007/978-981-10-6698-6_7