This project predicts which bank customers are likely to leave by analyzing demographic, financial, and behavioral data. Leveraging machine learning, I built a model that identifies at-risk customers, empowering the bank to implement retention strategies proactively. Through data preprocessing, EDA, and models like Random Forests, this solution enhances customer loyalty and drives profitability
Built a machine learning model to classify breast masses as malignant or benign based on FNA image features. Utilizing attributes like radius, texture, and compactness, models such as Random Forest and SVM were trained to support early detection, offering a valuable tool for medical diagnosis.
Developed interactive and insightful visualizations using Tableau and Power BI to uncover trends and patterns in complex datasets.
Showcasing expertise in SQL and NoSQL, these projects demonstrate my ability to manage and optimize both structured and unstructured data. From complex queries in SQL to scalable solutions with NoSQL databases like MongoDB, I deliver fast, flexible insights that drive impactful decisions.
This project performs an Exploratory Data Analysis (EDA) on a global terrorism dataset, with the goal of uncovering insights into terrorism trends, patterns, and outcomes worldwide. The analysis covers data cleaning, statistical analysis, and visualizations that provide insights into the nature and impact of terrorist attacks over time.
Analyzing over 3 million Instacart grocery orders, this project uncovers customer behavior patterns, performs segmentation for targeted marketing, and predicts future purchases. By exploring product associations and building machine learning models (XGBoost and Neural Network), I provide actionable insights for personalized recommendations and effective cross-selling, enhancing Instacart’s customer experience and sales strategies.
Objective of this project is to perform independent analysis of the wind-turbine data and predict the power curve of a wind turbine.