Welcome! This repository showcases a series of six hands-on projects developed as part of my academic training in data science and machine learning in the PGP AI and ML Program with UT Austin and Great Learning. Each folder contains a self-contained Jupyter notebook focused on a specific concept or application.
Below is an overview of the contents:
Course: Python Foundations
Perform an exploratory data analysis and provide actionable insights for a food aggregator company to get a fair idea about the demand of different restaurants and cuisines, which will help them enhance their customer experience and improve the business
Skills covered: python, numpy, pandas, seaborn, exploratory data analysis, business recommendations, bivariate analysis, univariate analysis
Course: Machine Learning
To identify bank customers with a high likelihood of purchasing a loan, you need to analyze the provided data to understand key customer attributes influencing loan acquisition. With this analysis, build a predictive model that captures patterns and customer characteristics, which will help the bank effectively target potential loan buyers, improving marketing efforts and increasing conversion rates.
Skills covered: exploratory data analysis, data pre-processing, model building, decision tree classifier, model performance evaluation and improvement, business recommendations
Course: Advanced Machine Learning
Analyze the data and come up with a predictive model to determine if a customer will leave the credit card services or not and the reason behind it
Skills covered: eda, random forest, bagging, boosting, smote, cross validation, data preprocessing, hyperparameter tuning
Course: Introduction to Neural Networks
Analyze the customer data, build a neural network to help the operations team identify the customers that are more likely to churn, and provide recommendations on how to retain such customers
Skills covered: eda, tensorflow, keras, artificial neural networks, regularization
Course: Introduction to Computer Vision
Build a robust image classifier using convolutional neural networks to efficiently classify different plant seedlings and weeds to improve crop yields and minimize human involvement
Skills covered: image processing, keras, tensorflow, convulational neural networks, transfer learning
Course: Introduction to Natural Language Processing
Develop an an AI-driven sentiment analysis system that will automatically process and analyze news articles to gauge market sentiment, and summarize the news at a weekly level to enhance the accuracy of their stock price predictions and optimize investment strategies.
Skills covered: large language models, text processing, transformers, prompt engineering, data manipulation, word embeddings, word2vec, glove
- All projects follow clean coding practices, include inline explanations, and use standard libraries (
scikit-learn,keras,seaborn, etc.). - Each notebook is designed to be understandable and reproducible.
Thanks for stopping by! ✨