A Natural Language Question Answering System as a Participant in Human Q&A Portals
Ingo Glöckner, Tiansi Dong, Ulrich Furbach and Björn Pelzer
LogAnswer is a question answering (QA) system for the German language, aimed at providing concise and correct answers to arbitrary questions. For this purpose LogAnswer is designed as an embedded artificial intelligence system which integrates methods and subsystems from several fields of AI, namely natural language processing, machine learning, knowledge representation and automated theorem proving. We intend to employ LogAnswer as a virtual user within internet-based QA forums. A key obstacle to gaining the acceptance of human visitors is the production of incorrect answers, as a virtual user will be perceived as a nuisance if its contributions are irrelevant or misleading. A QA system in this context must be able to identify the questions that it cannot answer correctly, a task that normally receives little attention in QA research compared to the actual answer derivation. Based on experiments with real-world data in the form of a large set of questions from a QA forum, we present filtering methods and machine learning features which significantly improve the avoidance of undesirable automated responses.