Logistic Regression Modeling with A Bayesian Approach to The Risk Factors of Colorectal Cancer Patients

Abstract

This research aims to create a model, determine the factors that affect the probability of someone suffering from colorectal cancer also design a web application to do prediction using the probability logit function which has been made. There are 4 independent variables and 1 dependent variable which are used in this research. The independent variables used in this research are heredity-pedigree value, age, Adenomatous Polyposis Coli, and MutS Homolog 2. And the dependent variable used in this research is someone’s status of suffering from colorectal cancer symbolized with Y. The stages of designing a web application include making use case diagrams, use case descriptions, activity diagrams, sequence diagrams, class diagrams, entity relationship diagrams, and designing the application interface. The results of this research will be a logit probability model which can be used to calculate the probability value of someone suffering from colorectal cancer. Model constructed using significant 3 independent variables which are heredity-pedigree, age, and Adenomatous Polyposis Coli.