A Composite Kernel to Extract Relations between Entities with Both Flat and Structured Features

Proceedings of the 21st International Conference on Computational Linguistics and 44th Annual Meeting of the ACL, pages 825–832, Sydney, July 2006. c⃝2006 Association for Computational Linguistics A Composite Kernel to Extract Relations between Entities with both Flat and Structured Features Min Zhang Jie Zhang Jian Su Guodong Zhou Institute for Infocomm Research 21 Heng Mui Keng Terrace, Singapore 119613 {mzhang, zhangjie, sujian, zhougd}@i2r.a-star.edu.sg Abstract This paper proposes a novel composite ker- nel for relation extraction. The composite kernel consists of two individual kernels: an entity kernel that allows for entity-related features and a convolution parse tree kernel that models syntactic information of relation examples. The motivation of our method is to fully utilize the nice properties of kernel methods to explore diverse knowledge for relation extraction. Our study illustrates that the composite kernel can effectively capture both flat and structured features without the need for extensive feature engineering, and can also easily scale to include more fea- tures. Evaluation on the ACE corpus shows that our method outperforms the previous best-reported methods and significantly out- performs previous two dependency tree ker- nels for relation extraction. 1 Introduction The goal of relation extraction is to find various predefined semantic relations between pairs of entities in text. The research on relation extrac- tion has been promoted by the Message Under- standing Conferences (MUCs) (MUC, 1987- 1998) and Automatic Content Extraction (ACE) program (ACE, 2002-2005). According to the ACE Program, an entity is an object or set of ob- jects in the world and a relation is an explicitly or implicitly stated relationship among entities. For example, the sentence “Bill Gates is chair- man and chief software architect of Microsoft Corporation.” conveys the ACE-style relation “EMPLOYMENT.exec” between the entities “Bill Gates” (PERSON.Name) and “Microsoft Corporation” (ORGANIZATION. Commercial). In this paper, we address the problem of rela- tion extraction using kernel methods (Schölkopf and Smola, 2001). Many feature-based learning algorithms involve only the dot-product between feature vectors. Kernel methods can be regarded as a generalization of the feature-based methods by replacing the dot-product with a kernel func- tion between two vectors, or even between two objects. A kernel function is a similarity function satisfying the properties of being symmetric and positive-definite. Recently, kernel methods are attracting more interests in the NLP study due to their ability of implicitly exploring huge amounts of structured features using the original represen- tation of objects. For example, the kernels for structured natural language data, such as parse tree kernel (Collins and Duffy, 2001), string ker- nel (Lodhi et al., 2002) and graph kernel (Suzuki et al., 2003) are example instances of the well- known convolution kernels1 in NLP. In relation extraction, typical work on kernel methods in- cludes: Zelenko et al. (2003), Culotta and Soren- sen (2004) and Bunescu and Mooney (2005). This paper presents a novel composite kernel to explore diverse knowledge for relation extrac- tion. The composite kernel consists of an entity kernel and a convolution parse tree kernel. Our study demonstrates that the composite kernel …

Content truncated to 500 words