How I won my 25th bronze medal on Kaggle’s Paris House Price competition

I woke up this morning to an email from Kaggle informing me that I had won my 25th bronze medal in their Playground competition, season 3 episode 6, which concerned house prices in Paris. I was able to achieve a high rank on the leaderboard by using XGBClassifier as the model for predicting house prices on the test set.

XGBClassifier is an implementation of gradient boosting trees for binary and multiclass classification problems. It builds an ensemble of decision trees to predict the target class for a given input. XGBoost handles missing values, categorical variables, and large datasets efficiently, and can handle imbalanced datasets by allowing you to specify the balance between positive and negative cases. The classifier can handle both dense and sparse data, and can be used in a variety of contexts, such as text classification, image classification, and speech recognition.

XGBClassifier uses the gradient boosting algorithm to train decision trees, where each tree is trained to correct the mistakes of the previous trees in the ensemble. The training process involves adjusting the weights of the instances, so that the mistakes of the previous trees are given more importance. This results in a set of trees that are able to capture the underlying patterns in the data and make accurate predictions.

I have written a blog post on this particular competition question and it can be viewed here:- https://medium.com/mlearning-ai/how-i-used-xgboost-to-make-predictions-on-a-paris-house-price-dataset-5d1c94b9df1c