Improving Information Extraction by Acquiring External Evidence with Reinforcement Learning
This work addresses the challenge of limited training data for information extraction, which is critical for domains like incident reporting and food safety, though it is incremental as it builds on existing reinforcement learning methods.
The paper tackles the problem of improving information extraction accuracy in data-scarce domains by acquiring external evidence through a reinforcement learning framework, resulting in significant performance gains over traditional extractors and a baseline meta-classifier on two databases.
Most successful information extraction systems operate with access to a large collection of documents. In this work, we explore the task of acquiring and incorporating external evidence to improve extraction accuracy in domains where the amount of training data is scarce. This process entails issuing search queries, extraction from new sources and reconciliation of extracted values, which are repeated until sufficient evidence is collected. We approach the problem using a reinforcement learning framework where our model learns to select optimal actions based on contextual information. We employ a deep Q-network, trained to optimize a reward function that reflects extraction accuracy while penalizing extra effort. Our experiments on two databases -- of shooting incidents, and food adulteration cases -- demonstrate that our system significantly outperforms traditional extractors and a competitive meta-classifier baseline.