Telecommunication Company Churn Project

Predictive model that tells important factors(or features) affecting the demand for shared bikes

Forecasting time series data using ARIMA models. Used covariance matrix to find dependencies between stocks.

Final Cybersecurity ML project of Marc Mestre and Yana Veitsman for Data Mining and Machine Learning course at University of Valencia, Spring 2021

Data Set: House Prices: Advanced Regression Techniques Feature Engineering with 80+ Features

Artificial Neural Network using Keras in python to identify customers who are likely to churn.

A Streamlit web app utilizing Python, scikit-learn, and pandas for used car price prediction. Features data preprocessing (scaling, encoding), Random Forest model optimization with GridSearchCV, and interactive user input handling. Achieves high accuracy (R² score: 0.9028), showcasing skills in machine learning, data engineering, and deployment.

This repository contains clustering techniques applied to minute weather data. It contains K-Means, Heirarchical Agglomerative clustering. I have applied various feature scaling techniques and explored the best one for our dataset

Wrangled real estate data from multiple sources and file formats, brought it into a single consistent form and analysed the results.

Feature transformation is a technique in machine learning that changes the way features are represented in order to improve the performance of machine learning algorithms. This can be done by transforming the features to a different scale, removing outliers, or creating new features from existing

This advanced forecasting tool leverages Prophet, ARIMA, SARIMA, and LSTM models to predict daily sales for 32 pizzas and 64 ingredients. With Prophet achieving the lowest MAPE, it ensures accurate demand forecasts, optimized inventory, and efficient purchase planning, reducing waste, preventing stockouts, and enhancing supply chain efficiency.

Build a machine learning model to predict if a credit card application will get approved.

This repository demonstrates the scaling of the data using Scikit-learn's StandardScaler, MinMaxScaler, and RobustScaler.

Bank Customer Behaviour Prediction

Exploratory Data Analysis for HR dataset

Time Series Model

Cloud image generation with Python and OpneCV