PMGC Winner Prediction Model

Overview

This project involved building a predictive model to determine the potential winners of the PUBG Mobile Global Championship (PMGC) based on historical performance data, team compositions, and map-specific statistics.

Key Features

• Data Collection: Scraped data from various e-sports databases.
• Feature Engineering: Created metrics like 'Consistency Index' and 'Clutch Factor'.
• Model Selection: Evaluated Random Forest, XGBoost, and Logistic Regression.
• Accuracy: Achieved a 78% prediction rate on the knockout stages.

Technologies Used

import sklearn
import pandas as pd
import numpy as np
# Primary model: XGBoost Classifier