Machine-Learning-Projects

Description of the Projects

• Project 1: Titanic - Machine Learning from Disaster

  • The project aims to utilize passenger data, including variables such as name, age, gender, socio-economic class, and more from the Titanic disaster to create a predictive model that can determine the likelihood of survival for passengers based on the provided features.

• Project 2: Biomechanical Features of Orthopedic Patients

  • In this project, we build a machine learning model based on the k-Nearest Neighbors algorithm with data of biomechanical features of orthopedic patients in order to classify patients as belonging to one out of three categories: Normal (100 patients), Disk Hernia (60 patients) or Spondylolisthesis (150 patients).

• Project 3: Weather Conditions in World War Two

  • In this project we are investigating the relationship between the minimum and maximum temperatures and utilizing a linear regression model for temperature prediction. Therefore, we use data from the "Weather Conditions in World War Two" dataset.

• Project 4: A/B-Testing

  • Given data that contains the advertisement click rate for two products, in this project we want to determine which product performs better. In order to achieve this goal, we use an A/B-Test.

• Project 5: House Prices - Advanced Regression/Feature Engineering Techniques

  • In this project, our emphasis is on feature engineering and advanced regression techniques using the house price dataset. Firstly, in our pursuit of feature engineering, we explore and discuss various methods to handle outliers and missing values, both in categorical and numeric columns. Secondly, in the context of advanced regression techniques, we apply different regression algorithms, such as a RandomForestRegressor.

• Project 6: Work on Time Series - Hourly Energy Consumption

  • In this project, we use hourly time-series data on energy consumption to predict future values for one year in advance. The model that we build in this project employs the XGBoost algorithm.