The exponential surge in healthcare data is providing new opportunities to discover meaningful data-driven characteristics and patterns of diseases. Machine learning and deep learning models have been employed for many computational phenotyping and healthcare prediction tasks. However, machine learning models are crucial for wide adaption in medical research and clinical decision making. In this chapter, the authors introduce demystifying diseases identification and diagnosis of various disease using machine learning algorithms like logistic regression, naive Bayes, decision tree, MLP classifier, random forest in order to cure liver disease, hepatitis disease, and diabetes mellitus. This work leverages the initial discoveries made through data exploration, statistical analysis of data to reduce the number of explanatory variables, which led to identifying the statistically significant attributes, mapping those significant attributes to the response, and building classification techniques to predict the outcome and demystify clinical diseases.
TopBackground
Machine Learning Algorithms are widely used in diagnosis of different clinical diseases. In recent years researchers adopted different machine learning classification algorithms for disease diagnosis and prediction.
Machine Learning Algorithms classified as supervised, Un-supervised and ensembles algorithms aims to focus on making predictions using computers in social media, financial, Medical, Entertainment and building product, movie and song recommendation engine domains. (Jafar Alzubi, Anand Nayyar & Akshi Kumar, 2018). In this chapter, we use various supervised learning algorithms like classification, regression and Ensemble techniques (Jafar Alzubi, Anand Nayyar & Akshi Kumar, 2018) such as Logistic Regression, decision tree, Naïve bayes and Random Forest Algorithms.
We build decision tree Classification algorithms and leaf node signifies final decision (Jafar Alzubi, Anand Nayyar & Akshi Kumar,2018) on various clinical diseases prediction and diagnosis.
We build Naïve Bayes Classifier which classifies based on Bayes theorem (Jafar Alzubi, Anand Nayyar & Akshi Kumar,2018). It is a statistical learning algorithm. It works on one inference that is the effect of an attribute value of a given class is independent of the values of the other attributes.