Web page Classifier using the DMOZ data.
(update) The DMOZ site is now in archive and hence you wont find the dataset . Although you may get it at Kaggle Kaggle DMOZ dataset with 5 classes only or you can use the DMOZ dataset that i had on my PC. DMOZ dataset with 12 classes
So this was my first project in Machine learning where i had to classify webpage by extracting text from the internet and predicting from the 12 classes. This comes under the hood of NLP.