Bank GoodCredit wants to predict cred score for current credit card customers. The cred score will denote a customer’s credit worthiness and help the bank in reducing credit default risk.

Practical Application 3. Assignment 17.1. Goal is to compare the performance of the classifiers, K Nearest Neighbor, Logistic Regression, Decision Trees, and Support Vector Machines utilizing a dataset related to marketing bank products over the telephone.

A research study on How do factors like alcohol consumption, age, ethnic background, and medical history affect the risk of developing Alzheimer's disease?

Developed a machine learning pipeline to predict customer churn with over 90% accuracy, leveraging data preprocessing, feature engineering, and Random Forest modelling. Conducted exploratory data analysis to uncover key drivers of churn, such as customer recency and cohorts from first transations.

A smart question tagging and matching system using traditional ML and NLP. It finds similar questions, tags them, and groups related ones to reduce duplicates. Built with TF-IDF, cosine similarity , KMeans & XgBoost, GBDT,Logistic Regression for fast, scalable, and real-world use.

Predicting passenger survival on the Titanic using an ensemble machine learning approach, achieving a Kaggle score of 0.77990. This project leverages stacking with Random Forest, Gradient Boosting, and SVM, enhanced by feature engineering and hyperparameter tuning, to model survival patterns effectively.

This project focuses on sentiment analysis of the NLTK movie reviews dataset using machine learning. I implemented and compared Multinomial Naive Bayes and Logistic Regression models after preprocessing text data with TF-IDF vectorisation. The goal was to classify reviews as positive or negative.

This project analyzes Twitter sentiment using NLP and Machine Learning. It preprocesses text, converts it into numerical format, and trains a Logistic Regression model. The model classifies tweets as Positive, Negative, or Neutral and is evaluated using accuracy metrics. It can also predict the sentiment of new user-input statements.