This project involves the development of predictive models to predict survival on the Titanic. It includes thorough data exploration, preprocessing, and training of Random Forest and Gradient Boosting classifiers. The process also involves fine-tuning of hyperparameters, with the final predictions saved as CSV files.

This project analyzes the effect of weather data on stock market performance. It involves data collection, preprocessing, and analysis to uncover patterns and insights related to weather conditions and stock market trends.

The Agile Code Quality Assessment Tool automates code quality evaluation with advanced machine learning. It features a Quality Rating System and Detailed Issue Reporting for actionable insights. With adaptability to various coding standards, it's an essential tool for Agile developers seeking to optimize code quality and streamline review processes.

Embark on a data-driven journey to predict term deposit subscriptions with our comprehensive notebook. Explore dataset nuances and decode variable contributions for informed decision-making. From logistic regression to decision trees, our models offer validation accuracies exceeding 90%, providing robust predictions.

In this project, we delve into a comprehensive analysis of historical temperature data spanning over a century in India. Utilizing Plotly for visualization, we uncover intriguing insights into temperature fluctuations over time, identifying seasonal patterns and long-term trends.

In this project, we explore the MNIST dataset, consisting of hand-written digits, to develop accurate digit recognition models. Leveraging TensorFlow and Keras, we construct Convolutional Neural Networks (CNNs) with varying configurations, including different numbers of fully connected layers.

Financially, the store is doing reasonably well, with total revenue of $4.72 million and a healthy margin of $472,360. That's a solid performance, though there's always room for improvement.

This project involved analyzing a Telco customer dataset to understand factors influencing churn. Initial exploration revealed imbalanced data, with 73% not churning. Data cleaning included handling missing values and converting categorical variables. Insights emerged from univariate and bivariate analysis, showing higher churn among month-to-month contract holders.