This is the repository used for the final project of Text Analytics module from Uni of Essex
Here you can find a project related to XMLC (Extreme Multi-label classification) of text based on titles. The datasets are large so they have not been included in this repository, but they can be found in https://www.kaggle.com/hsrobo/titlebased-semantic-subject-indexing
Two approaches have been used: Classical Supervised Learning Algorithm (One-vs-All) and Deep Learning. You can find their respective code within each folder.
Also, there is a report that includes the results achieved by each method and their comparison to the most similar academic paper.