Open Domain Question Answering Using Web Tables

TL;DR

This paper presents a novel open-domain question answering system that leverages web tables to answer both factoid and non-factoid queries, significantly improving accuracy and deployed at scale.

Contribution

It introduces a combined neural and feature-based approach for web table QA that handles diverse query types and outperforms existing methods.

Findings

Outperforms state-of-the-art baselines in real-world web search queries

Successfully deployed in a commercial search engine

Serves tens of millions of user queries monthly

Abstract

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.