Open AccessThyroid Disease Prediction Using XGBoost Algorithms
Author(s) -
Sharmila Sankar,
Anupama Potti,
G. Naga Chandrika,
Somula Ramasubbareddy
Publication year - 2022
Publication title -
journal of mobile multimedia
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.229
H-Index - 12
eISSN - 1550-4654
pISSN - 1550-4646
DOI - 10.13052/jmm1550-4646.18322
Subject(s) - decision tree , computer science , logistic regression , disease , machine learning , statistical classification , algorithm , artificial intelligence , k nearest neighbors algorithm , thyroid function , thyroid disease , thyroid , data mining , medicine
Nowadays, thyroid disease is increasing rapidly all over the world. Significantly, one out of ten people is affected by the thyroid in India. In recent years, many researchers have done various research works on thyroid disease detection. Therefore, the early stage of thyroid disease prediction is difficult to protect and avoid the worst health condition. In this regard, the machine learning plays a crucial role to detect the disease accurately. We consider the UC Irvin knowledge discovery dataset. So, this paper proposes the XGBoost algorithm to predict thyroid disease accurately. The best features are selected using XGBoost function. The proposed XGBoost algorithm’s efficacy is compared to decision tree, logistic regression, k-Nearest Neighbor (kNN) methods. The performance of all four algorithms is compared and analyzed. It is observed that the accuracy of the XGBoost algorithm increases by 2% than the KNN algorithm.