UTD: Classifying Semantic Relations by Combining Lexical and Semantic Resources

Proceedings of the 5th International Workshop on Semantic Evaluation, ACL 2010, pages 256–259, Uppsala, Sweden, 15-16 July 2010. c⃝2010 Association for Computational Linguistics UTD: Classifying Semantic Relations by Combining Lexical and Semantic Resources Bryan Rink and Sanda Harabagiu Human Language Technology Research Institute University of Texas at Dallas Richardson, Texas {bryan,sanda}@hlt.utdallas.edu Abstract This paper describes our system for SemEval-2010 Task 8 on multi-way clas- sification of semantic relations between nominals. First, the type of semantic re- lation is classified. Then a relation type- specific classifier determines the relation direction. Classification is performed us- ing SVM classifiers and a number of fea- tures that capture the context, semantic role affiliation, and possible pre-existing relations of the nominals. This approach achieved an F1 score of 82.19% and an ac- curacy of 77.92%. 1 Introduction SemEval-2010 Task 8 evaluated the multi-way classification of semantic relations between nom- inals in a sentence (Hendrickx et al., 2010). Given two nominals embedded in a sentence, the task requires identifying which of the fol- lowing nine semantic relations holds between the nominals: Cause-Effect, Instrument-Agency, Product-Producer, Content-Container, Entity- Origin, Entity-Destination, Component-Whole, Member-Collection, Message-Topic, or Other if no other relation is appropriate. For instance, the following sentence provides an example of the Entity-Destination relation: “A small [piece]E1 of rock landed into the [trunk]E2.” The two nominals given for this sentence are E1 (piece) and E2 (trunk). This is an Entity- Destination relation because the piece of rock originated from outside of the trunk, but ended up there. Finally, the direction of the relation is (E1,E2) because E1, the piece, is the Entity and E2, the trunk, is the Destination. Analysis of the training data revealed three ma- jor classes of knowledge required for recognizing semantic relations: (i) examples that require back- ground knowledge of an existing relation between the nominals (e.g., example 5884 below), (ii) ex- amples using background knowledge regarding the typical role of one of the nominals (e.g., ex- ample 3402), and (iii) examples that require con- textual cues to disambiguate the role between the nominals (e.g., example 5710). Example 5884 “The Ca content in the [corn]E1 [flour]E2 has also a strong dependence on the pericarp thickness.” Example 3402 “The [rootball]E1 was in [crate]E2 the size of a refrigerator, and somea of the arms were over 12 feet tall.” Example 5710 “The seniors poured [flour]E1 into wax [paper]E2 and threw the items as projectiles on freshmen during a morning pep rally.” In example 5884, the background knowledge that flour is often made or derived from corn can directly lead to the classification of the example as containing an Entity-Origin relation. Likewise, knowing that crates often act as containers is a strong reason for believing that example 3402 is a Content-Container relation. However, in exam- ple 5710, neither the combination of the nominals nor their individual affiliations lead to an obvious semantic relation. After taking the context into account, it becomes clear that this is an Entity- Destination relation because E1 is going into E2. 2 Approach We …

Content truncated to 500 words