Machine Learning based Disease Prediction Model

Course: Supervised Learning - Classification

Built a binomial machine learning model to predict Parkinson's disease. The objective of this model is to classify patients with Parkinson's disease. Performed the required pre-processing steps prior to model building and evaluated the model performance through a proper tuning process.

Skills & Tools Covered:

• Machine Learning
• Supervised Learning
• Linear and Non-Linear Classification Models
• Hyperparameter Tuning
• Model Metrics

Machine Learning Model Building for Continuous Variable Prediction

Course: Supervised Learning - Regression

Built a Linear Machine Learning model to understand the relationship between the population of US cities in the years 1920 and 1930. Evaluated the model performance with appropriate measures. Performed all required graphical and quantitative exploratory data analysis prior to model building. The dataset comprises 49 rows and 2 columns, representing the population (in thousands) of 49 U.S. cities in 1920 and 1930. These cities are a random sample from the 196 largest cities in the U.S.

Skills & Tools Covered:

• Machine Learning
• Supervised Learning
• Linear Regression

Descriptive Statistics on Real World Problem

Course: Statistics for Machine Learning

This project involves tackling three real-world problem statements:

• The first problem focuses on exploratory data analysis of a La Liga Cup (a football tournament). The goal is to analyze the data for insights into top teams, winning distributions, and more.
• The second problem helps in understanding and applying confidence intervals to a healthcare-related issue.

Skills & Tools Covered:

• Descriptive and Inferential Statistics
• Hypothesis Testing
• Confidence Interval
• Distribution