publications thesis contact

Web-shop Order Prediction Using Machine Learning

In the prudsys Data Mining Cup 2013, teams build classifiers to predict whether a web-shop user session will end in an order.

We apply a knowledge discovery framework, review various approaches for dealing with missing data, evaluate prediction models, and finally build a stacking classifier consisting of support vector machine, Random Forest and neural network base classifiers using the R environment.


machine learning; e-commerce; imputation; stacked generalization; Random Forest; support vector machines; neural networks