Python Assignment: Predicting Education Level of Party Candidates

Introduction

This project implements a machine learning model to predict the education level of party candidates based on features such as Party, Criminal Cases, State, Liabilities, and more.

Libraries Used

• NumPy
• Pandas
• Matplotlib
• Scikit-learn (specifically, the RandomForestClassifier from the ensemble module)

Pre-Processing Data

Before training the machine learning model, the data undergoes preprocessing steps:

• Label Encoding: Categorical variables in the dataset are encoded using the LabelEncoder from scikit-learn. This is done for both the target variable (Education) and the features (Party, Criminal Case, State).

Model Selection and Training

The project involves tuning a RandomForestClassifier using cross-validation. Here's a brief description of the process:

1. Train-Test Split: The dataset is split into training and validation sets using train_test_split. This allows training the model on one subset of the data and evaluating its performance on another.
2. Evaluation Function: A function called get_mae is defined, which takes the maximum number of leaf nodes (max_leaf_nodes) as a parameter. Inside this function, a RandomForestClassifier model is trained with the specified max_leaf_nodes to compute the mean absolute error (MAE) between the actual and predicted values on the validation set.

Note: The final submission does not use max_leaf_nodes due to decreased performance with this parameter.

Plots

1. Percentage distribution of parties with candidates having the most criminal records:

   

2. Percentage distribution of parties with the most wealthy candidates:

   

Results

• Public F1 score: 0.24282
• Private F1 score: 0.21466

References

• Kaggle Tutorial given in Discussion section.