Open Domain Question Answering Using Web Tables
It addresses the limitation of previous works focused only on factoid queries, enabling broader application in web search for real users.
The paper tackled the problem of open-domain question answering using web tables for both factoid and non-factoid queries, resulting in a method that significantly outperforms state-of-the-art baselines and serves tens of millions of queries per month in a commercial search engine.
Tables extracted from web documents can be used to directly answer many web search queries. Previous works on question answering (QA) using web tables have focused on factoid queries, i.e., those answerable with a short string like person name or a number. However, many queries answerable using tables are non-factoid in nature. In this paper, we develop an open-domain QA approach using web tables that works for both factoid and non-factoid queries. Our key insight is to combine deep neural network-based semantic similarity between the query and the table with features that quantify the dominance of the table in the document as well as the quality of the information in the table. Our experiments on real-life web search queries show that our approach significantly outperforms state-of-the-art baseline approaches. Our solution is used in production in a major commercial web search engine and serves direct answers for tens of millions of real user queries per month.