College of Computer Science and Mathematics

Discussion of the master’s thesis in the College of Computer Science and Mathematics -Department of Statistics and Informatics Department entitled:

Using Cluster Analysis to Analyze Affecting Factors with Forecasting

Master’s thesis by  Abdulsalam Idres Abdulkarim

supervised by  Dr.Wisam Wadallah Saleem

This thesis includes

A study on type 2 diabetes involving data from (237) individuals, both affected and unaffected, collected from the Community Health Department / Nineveh Health Directorate.

The study addressed

The study utilizes Hierarchical Cluster Analysis to classify the most influential factors affecting diabetes. Following this, the best regression equation for the variables within each cluster is predicted by applying logistic regression. This approach aims to identify the most impactful variables and accurately predict the risk of diabetes with high precision. The entire analysis is implemented using the (Python 3)

The study aims

The study focuses on predicting the most influential factors affecting diabetes by integrating the results of cluster analysis with logistic regression. This combined approach is used to analyze the impact of various factors on the target variables and predict the most significant factors contributing to diabetes. An improved logistic regression model is developed using the extracted clusters, enabling highly accurate predictions of the dependent variables. This methodology ensures precise and reliable results in forecasting diabetes-related outcomes.