A Comprehensive Comparison of Machine Learning Classification Techniques for Early Detection of Alzheimer's Disorder

Alzheimer's disorder (AlzD) is a chronic condition in which brain cells degenerate and eventually die. People over the age of 65 are most likely to get this condition. There is no cure for this, but nowadays proper medications and treatments are available to slow down the disease progression temporarily. Detection of AlzD in its early stage may help doctors and physicians provide proper medicines as early as possible to slow its development. Artificial intelligence is a blooming field in every sector nowadays that can help in AlzD detection to save us from time-consuming and expensive conventional methods. Various machine learning algorithms are researched to prevent AlzD. This paper uses different linear, distance, and tree-based supervised classifier algorithms. The used dataset contains different psychological parameters like a person's age, number of visits, Mini-Mental State Examination (MMSE), and years of education, on which machines can be trained to predict whether the person is demented or non-demented. The main goal of this research paper is to compare these classification algorithms which give the best result in the early detection of AlzD. RFC, KNN, and MLP algorithms outperform all other algorithms with an accuracy of 96.77<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup>, followed by NB and LR algorithms with 95.70%.