Fast and Large-scale Unsupervised Relation Extraction

Fast and Large-scale Unsupervised Relation Extraction Sho Takase† Naoaki Okazaki†‡ Kentaro Inui† Graduate School of Information Sciences, Tohoku University† Japan Science and Technology Agency (JST)‡ {takase, okazaki, inui}@ecei.tohoku.ac.jp Abstract A common approach to unsupervised relation extraction builds clusters of patterns express- ing the same relation. In order to obtain clus- ters of relational patterns of good quality, we have two major challenges: the semantic rep- resentation of relational patterns and the scal- ability to large data. In this paper, we ex- plore various methods for modeling the mean- ing of a pattern and for computing the similar- ity of patterns mined from huge data. In order to achieve this goal, we apply algorithms for approximate frequency counting and efficient dimension reduction to unsupervised relation extraction. The experimental results show that approximate frequency counting and dimen- sion reduction not only speeds up similarity computation but also improves the quality of pattern vectors. 1 Introduction Semantic relations between entities are essential for many NLP applications such as question an- swering, textual inference and information extrac- tion (Ravichandran and Hovy, 2002; Szpektor et al., 2004). Therefore, it is important to build a compre- hensive knowledge base consisting of instances of semantic relations (e.g., authorOf) such as authorOf ⟨Franz Kafka, The Metamorphosis⟩. To recognize these instances in a corpus, we need to obtain pat- terns (e.g., “X write Y”) that signal instances of the semantic relations. For a long time, many researches have targeted at extracting instances and patterns of specific rela- tions (Riloff, 1996; Pantel and Pennacchiotti, 2006; De Saeger et al., 2009). In recent years, to acquire a wider range knowledge, Open Information Extrac- tion (Open IE) has received much attention (Banko et al., 2007). Open IE identifies relational patterns and instances automatically without predefined tar- get relations (Banko et al., 2007; Wu and Weld, 2010; Fader et al., 2011; Mausam et al., 2012). In other words, Open IE acquires knowledge to han- dle open domains. In Open IE paradigm, it is nec- essary to enumerate semantic relations in open do- mains and to learn mappings between surface pat- terns and semantic relations. This task is called unsupervised relation extraction (Hasegawa et al., 2004; Shinyama and Sekine, 2006; Rosenfeld and Feldman, 2007). A common approach to unsupervised relation ex- traction builds clusters of patterns that represent the same relation (Hasegawa et al., 2004; Shinyama and Sekine, 2006; Yao et al., 2011; Min et al., 2012; Rosenfeld and Feldman, 2007; Nakashole et al., 2012). In brief, each cluster includes patterns corre- sponding to a semantic relation. For example, con- sider three patterns, “X write Y”, “X is author of Y” and “X is located in Y”. When we group these patterns into clusters representing the same relation, patterns “X write Y” and “X is author of Y” form a cluster representing the relation authorOf, and the pattern “X is located in Y” does a cluster for locate- dIn. In order to obtain these clusters, we need to know the similarity between patterns. The …

Content truncated to 500 words