Stack Exchange Tagger

TL;DR

This paper presents a multilabel classification approach for tagging Stack Exchange questions, comparing various SVC configurations and finding linear SVC with Crammer Singer to be most effective.

Contribution

It introduces an effective multilabel classifier for Stack Exchange questions and evaluates different SVC configurations for optimal performance.

Findings

Linear SVC with Crammer Singer yields best results

Support Vector Classification is effective for large-scale multilabel text classification

Comparison of kernel functions and loss functions for SVC

Abstract

The goal of our project is to develop an accurate tagger for questions posted on Stack Exchange. Our problem is an instance of the more general problem of developing accurate classifiers for large scale text datasets. We are tackling the multilabel classification problem where each item (in this case, question) can belong to multiple classes (in this case, tags). We are predicting the tags (or keywords) for a particular Stack Exchange post given only the question text and the title of the post. In the process, we compare the performance of Support Vector Classification (SVC) for different kernel functions, loss function, etc. We found linear SVC with Crammer Singer technique produces best results.