G

gene-expression

This project classifies brain cancer using gene expression data. It includes preprocessing, class balancing, feature selection, and training models (Random Forest, XGBoost, SVM) in pipelines to compare performance to better insights in biomedical informatics and cancer research.

This is a capstone group project for the Data Mining course CSCE 874 at UNL. There were 4 team members. My contribution in this project lies in data shaping using techniques like binarization, and discretization. Then I used association rule mining to predict maize yield by connecting maize's phenotypic characters such as cob length, mass, ear length, etc., and the available gene expressions.