Knowledge-guided Open Attribute Value Extraction with Reinforcement Learning

Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, pages 8595–8604, November 16–20, 2020. c⃝2020 Association for Computational Linguistics 8595 Knowledge-guided Open Attribute Value Extraction with Reinforcement Learning Ye Liu1∗ Sheng Zhang1∗ Rui Song1† Suo Feng2 Yanghua Xiao2† 1Department of Statistics, North Carolina State University, Raleigh, NC 27695-8206 2Shanghai Key Laboratory of Data Science, School of Computer Science, Fudan University, China 200433 {yliu87, szhang37, rsong}@ncsu.edu, {fengs17, shawyh}@fudan.edu.cn Abstract Open attribute value extraction for emerging entities is an important but challenging task. A lot of previous works formulate the problem as a question-answering (QA) task. While the collections of articles from web corpus pro- vide updated information about the emerging entities, the retrieved texts can be noisy, ir- relevant, thus leading to inaccurate answers. Effectively filtering out noisy articles as well as bad answers is the key to improving ex- traction accuracy. Knowledge graph (KG), which contains rich, well organized informa- tion about entities, provides a good resource to address the challenge. In this work, we pro- pose a knowledge-guided reinforcement learn- ing (RL) framework for open attribute value extraction. Informed by relevant knowledge in KG, we trained a deep Q-network to sequen- tially compare extracted answers to improve extraction accuracy. The proposed framework is applicable to different information extrac- tion system. Our experimental results show that our method outperforms the baselines by 16.5 - 27.8%. 1 Introduction Numerous entities are emerging everyday. The attributes of the entities are often noisy or incom- plete, even missing. In the field of electronic com- merce, target attributes (e.g., brand, flavor, smell) of new products are often missing (Zheng et al., 2018). In medical analysis, attributes like trans- mission, genetics and origins of a novel virus are often unknown to people. Even in DBpedia, a well-constructed and large-scale knowledge base extracted from Wikipedia, half of the entities con- tain less than 5 relationships (Shi and Weninger, ∗Ye Liu and Sheng Zhang contributed equally. †Rui Song and Yanghua Xiao are corresponding authors. Yanghua Xiao was supported by Shanghai Science and Tech- nology Innovation Action Plan (No.19511120400). 2018). A method that is capable of supplementing reliable attribute values for emerging entities can be highly useful in many applications. Although information extraction methods have been extensively studied, the task of open attribute value extraction remains challenging. First, the emerging entities may have new attribute values that are absent in the existing KG. Under such circumstances, the prediction methods under the closed-world assumption and the methods that can- not utilize external information are not well suited due to their limited recalls. Second, while web corpus can be used as a good resource to provide relatively updated and relevant articles for large varieties of emerging entities, the articles retrieved from web corpus can be noisy and/or irrelevant, which in turn leads to a limited precision. Finally, even when articles are relevant, the extracted an- swers might still be inaccurate due to the error- prone information extraction model. To effectively filter out noisy answers that are obtained …

Content truncated to 500 words