House Prices
At the top, the distribution of starting and normalized samples. Below, scatter plots of two features.
Box plot diagrams of some features.
On the left, a correlation matrix; on the right, the Gini importance of features.
Summary table of the results from various models. At the bottom, the score and ranking in the competition.
Type: academic project
Technologies: Python, Kaggle, Visual Studio Code
Notebook created for a Kaggle competition, aimed at predicting house prices in a given dataset based on their features and a predefined training set. The analysis included data exploration, pre-processing, and the selection of advanced regression models, which were compared using appropriate metrics. The final result placed the project in 100th position, ranking in the top 3% of participants.