0. Machine Learning Preparation Projects

JANUARY TICKET: 133AB8B53396F51B392B

This course is a curated collection of both Supervised and Unsupervised Machine Learning content.

IMPORTANT: This course reuses material from two previous courses available on my Udemy profile. If you have already taken them, note that this is a combined and reorganized version, designed to offer a unified learning experience without switching between separate courses.

5. Practical Projects and Real Applications

Project 1: A/B Testing on Web Traffic

In this project, you’ll work with real web traffic data from a company to evaluate the impact of two page versions on user conversion rates. You will learn to:

• Clean and explore traffic data using Pandas.

• Apply statistical tests like the t-test and hypothesis testing to compare conversion rates.

• Visualize results and make data-driven decisions.

• Create a report with actionable insights to optimize website performance.

Project 2: Bank Data Exploration for Machine Learning

In this practical case, you’ll analyze a financial dataset containing bank client information to detect patterns and prepare data for future Machine Learning models. You’ll focus on:

• Data cleaning and transformation using Pandas and NumPy.

• Visual exploration with Seaborn to identify trends and correlations.

• Using descriptive statistics to understand client behavior.

• Preparing the dataset for predictive modeling.

Project 3: Insurance Data Analysis

You will work with insurance company data to identify key factors affecting policy costs and client risk. You will learn to:

• Clean and transform customer and claim data.

• Apply EDA techniques to find trends and anomalies.

• Use descriptive stats and visualizations to support decisions.

• Prepare data for predictive risk models.

Project 4: Malware Data Analysis

In this project, you will analyze a dataset containing malware information to detect cyberattack patterns and help prevent vulnerabilities. You will focus on:

• Cleaning and structuring malware data.

• Visual exploration to identify relevant features.

• Applying preprocessing techniques to improve data quality.