A Machine Learning Approach For Factoid Question Answering

This paper presents a factoid Question Answering system that is fully based on machine learning. Our system achieves similar results to a state-of-the-art QA system with answer extraction rules developed by a human expert. Our approach avoids human intervention and simplifies adaptation of the system to new environments or extended feature sets. Moreover, its response time is suitable for places where real-time Question Answering is needed.