Type Checking in Open-Domain Question Answering

Open domain question answering (QA) systems have to bridge the potential vocabulary mismatch between a question and its candidate answers. One can view this as a recall problem and address it accordingly. Recall oriented strategies to QA may generate considerable amounts of noise. To combat this, many open domain QA systems contain an explicit filtering or re-ranking component, which often check whether the answer is of the correct semantic type. Particular classes of questions expect specific answer types to which all of their answers should belong. We compare two kinds of strategies for answer type checking for open domain QA. One is redundancy-based and builds on the intuition that the amount of implicit knowledge which connects an answer to a question can be estimated by exploiting the redundancy of information available on the web. The other is knowledge-intensive, and exploits structured and semi-structured data sources to determine, with high confidence, the semantic type of suggested answers.