Data Preprocessing
- Added data preprocessing pipeline (encoding, scaling, train-test split)
- Handled missing values and outliers for better model performance
- Engineered new features (Efficiency_Per_Hour, Meeting_Impact)
Exploratory Data Analysis (EDA)
- Plotted correlation matrix & pairplots for insights
- Visualized productivity distribution and key features
- Analyzed impact of meetings & work-life balance on productivity
Machine Learning Models
- Implemented Random Forest Classifier for productivity categories
- Trained regression model to predict Productivity_Score
- Tuned hyperparameters for better model accuracy
Evaluation & Optimization
- Generated classification report & accuracy metrics
- Computed R², MAE, RMSE for regression performance
- Optimized feature selection for improved model performance