Feature Engineering for Knowledge Base Construction

TL;DR

This paper discusses a feature engineering approach for knowledge base construction using probabilistic inference, emphasizing the importance of features over algorithms, and introduces the DeepDive system to facilitate this process.

Contribution

It presents a novel feature engineering methodology for KBC and introduces DeepDive, a system that enables systematic, loosely coupled knowledge base construction.

Findings

Knowledge bases built with this approach match or surpass human quality.

DeepDive allows construction of knowledge bases by a single graduate student.

Feature engineering is crucial for end-to-end quality in KBC.

Abstract

Knowledge base construction (KBC) is the process of populating a knowledge base, i.e., a relational database together with inference rules, with information extracted from documents and structured sources. KBC blurs the distinction between two traditional database problems, information extraction and information integration. For the last several years, our group has been building knowledge bases with scientific collaborators. Using our approach, we have built knowledge bases that have comparable and sometimes better quality than those constructed by human volunteers. In contrast to these knowledge bases, which took experts a decade or more human years to construct, many of our projects are constructed by a single graduate student. Our approach to KBC is based on joint probabilistic inference and learning, but we do not see inference as either a panacea or a magic bullet: inference is a tool that allows us to be systematic in how we construct, debug, and improve the quality of such systems. In addition, inference allows us to construct these systems in a more loosely coupled way than traditional approaches. To support this idea, we have built the DeepDive system, which has the design goal of letting the user "think about features---not algorithms." We think of DeepDive as declarative in that one specifies what they want but not how to get it. We describe our approach with a focus on feature engineering, which we argue is an understudied problem relative to its importance to end-to-end quality.