Kaggle Contests - Machine Learning Experiments

This repository contains my solutions, experiments that I performed in the Kaggle Competitions that I had participated in. My goal by participating in these Kaggle Competitions was to level up my understanding related to Core Machine Learning Algorithms and Data Pre processing and various edge cases that we need to handle ususally.

Competitions Covered: 1.) Predicting Road Accident Risk (https://www.kaggle.com/competitions/playground-series-s5e10)

A contest which involved me training Machine Learning models to predict the Likelihood of accidents on different types of roads using features such as

Road curvature and number of lanes
Weather and lighting conditions
Presence of road signs
Holiday or school season flags
Speed limits, time of day, and more

2.) Predicting the Beats-per-Minute Of Songs (https://www.kaggle.com/competitions/playground-series-s5e9)

A contest related to developing Machine learning solutions to predict the BeatsPerMinute based on Music Track BPM data. The features that I worked with this in contest are as listed below

RhythmScore and AudioLoudness
VocalContent and AcousticQuality
InstrumentalScore and LivePerformance
MoodScore and TrackDurationMs
Energy

3.) Predicting Loan Payback (https://www.kaggle.com/competitions/playground-series-s5e11/data)

A contest where I had to develop Machine Learning Solutions to predict if the given customer with the give features will be able to payback the loan or not. The features that I worked with are listed below

1. Borrower’s Demographics

  age (int64) – Borrower's age (in years).

  gender (category) – Borrower's gender (Male/Female).

  marital_status (category) – Marital status (Single, Married, Divorced).

  education_level (category) – Education level (High School, Bachelor, Master, PhD).

2. Financial Information

   annual_income (float64) – Borrower's yearly income.

   monthly_income (float64) – Borrower's monthly income.

   employment_status (category) – Current employment type (Employed, Self-Employed, Unemployed).

   debt_to_income_ratio (float64) – Ratio of borrower’s debt to their income. Lower = better.

  credit_score (int64) – Credit bureau score (e.g., FICO). Higher = less risky.

3. Loan Information

   loan_amount (float64) – Amount of loan taken.

   loan_purpose (category) – Loan purpose (Car, Education, Home, Medical, etc.).

   interest_rate (float64) – Loan par annual interest rate (%).

   loan_term (int64) – Loan repayment duration (months, e.g., 36 or 60).

   installment (float64) – Monthly installment .

   grade_subgrade (category) – Risk category assigned to loan (A1, B2, etc.).
4. Borrower’s Credit History

   num_of_open_accounts (int64) – Total active credit accounts.

   total_credit_limit (float64) – Borrower's total available credit limit.

   current_balance (float64) – Borrower's outstanding balance (loan + credit card).

   delinquency_history (int64) – Count of late payments in borrower’s history.

   public_records (int64) – Negative public records (e.g., bankruptcies, legal actions).

   num_of_delinquencies (int64) – Total delinquencies (missed payments).

Prequisites -Python 3.8+

Setup Clone the repository: git clone https://github.com/Zrahay/Kagg-Repo.git