You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Discover a comprehensive approach to constructing credit risk models. We employ various machine learning algorithms like LightGBM and CatBoost, alongside ensemble techniques for robust predictions. Our pipeline emphasizes data integrity, feature relevance, and model stability, crucial elements in credit risk assessment.
We all have experienced a time when we have to look up for a new house to buy. But then the journey begins with a lot of frauds, negotiating deals, researching the local areas and so on. So to deal with this kind of issues Today, I prepared a MACHINE LEARNING Based model, trained on the House Price Prediction Dataset.
The Credit Card Fraud Detection Problem includes modeling past credit card transactions with the knowledge of the ones that turned out to be a fraud. This model is then used to identify whether a new transaction is fraudulent or not. Our aim here is to detect 100% of the fraudulent transactions while minimizing the incorrect fraud classifications.
This project develops an advanced predictive model to identify thyroid disease recurrence using machine learning algorithms. We used a detailed dataset with demographic, medical, and clinical features, and implemented Logistic Regression, Decision Tree, Random Forest, and CatBoost Classifier. Rigorous preprocessing and EDA were performed.
A Domestic violence support system for the victims, that enables users to share their thought and provides knowledge about the particular type of abuse they are going through.
Ad huc solution for anomaly classification of HTTP requests between service and end-user based on limited data / Решение задачи поиска аномальных HTTP запросов (их классификации) к сервису.
The purpose is to train a predictive model that can determine if a given customer will subscribe to a term deposit based on these various features. By analyzing historical data on successful and unsuccessful subscription outcomes, patterns can be identified which help predict future subscription behavior.
Uses letter frequency and catboost classifier model in synchronous for guessing letters in hangman game instance. The model performance is evaluated on both seen words in the dictionary and words out of the dictionary.
Kaggle competition on network intrusion detection. Train model, predict test set, submit as CSV (ID, Class). F1-score metric. Part of my 2023 master's program at the University of Ottawa in AI for Cyber Security.