Cancer Cell Sensitivity Based on Gene Expression & Machine Learning

Processed and analyzed complex genomic datasets using Python and machine learning techniques.

● Applied data preprocessing and exploratory analysis to prepare high-dimensional biological data for modeling.

● Developed a LightGBM Regressor (leaf-wise gradient boosting) to model relationships between gene expression and drug

sensitivity, well-suited to large-scale, high-dimensional genomic data.

● Improved model accuracy from 80.57% (before normalization) to 95.04% (after normalization) through feature

normalization.