Abstract
We empirically investigate the task of classifying adjectives into property-denoting vs. relational types, a distinction that is highly relevant for ontology learning. The feasibility of this task is evaluated in two experiments: (i) a corpus study based on human annotations and (ii) an automatic classification experiment. We observe that token-level annotation of these classes is expensive and difficult. Yet, a careful corpus analysis reveals that adjective classes tend to be stable on the type level, with few occurrences of class shifts observed at the token level. As a consequence, we opt for an automatic classification approach that operates on the type level. Training on heuristically labeled data yields high classification performance on our own data and on a data set compiled from WordNet. Our results indicate that it is feasible to automatically distinguish property-denoting and relational adjectives, even if only small amounts of annotated data are available. A combination of semantic, morphological and shallow syntactic features turns out to be most informative for the task.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
A strict comparison of the two approaches will not be possible due to the different languages considered and divergences regarding the selected target classes.
- 2.
We used version 3 of the BNC XML Edition, available from: http://www.natcorp.ox.ac.uk/
- 3.
Part of speech tagging was the primary source of errors here.
- 4.
In its original statement, the notion of foundation is defined as follows: “For a concept α to be founded on another concept β, any instance χ of α has to be necessarily associated to an instance ϕ of β which is not related to χ by a part-of relation” (Guarino, 1992). We extend this notion from concepts to properties, arguing that event-based adjectives denote founded properties that are necessarily associated with an implicit event.
- 5.
Note that these patterns are mutually exclusive: (5a) applies to examples such as comfortable chair and interesting article in (2b) and (2c), where ENT fills the patient role of EVENT. In contrast, eloquent person in (2a) can be identified as event-based by (5b) only, as ENT acts as the agent of EVENT here (cf. Lapata, 2001). We expect that disambiguating basic and event-related readings should work best if (5a) is constrained such that EVENT may not be instantiated by perception verbs such as look, feel, taste etc.
- 6.
κ measures the agreement among annotators on classification tasks. Its values range between 0 (no agreement at all) and 1 (perfect agreement), reflecting the degree of agreement above chance (Fleiss, 1971).
- 7.
Henceforth, we will refer to these binary classes as ATTR(ibutive) and REL(ational).
- 8.
In a selective investigation on more representative data, class volatility turns out to be only slightly higher (cf. Sect. 8.6.1).
- 9.
The abbreviations used in the table denote part-of-speech tags according to the Penn Treebank nomenclature (Marcus et al., 1993).
- 10.
Note that the pertainymy relation in WordNet is uni-directional as it contains only links from adjectives to their morphological base nouns, but not from derived nouns to base adjectives. For instance, cultural and culture or dental and tooth are linked by pertainymy, while no such link exists between short and shortness.
References
Almuhareb, Abdulrahman. 2006. Attributes in lexical acquisition. Ph.D. dissertation, Department of Computer Science, University of Essex.
Almuhareb, Abdulrahman, and Massimo Poesio. 2004. Attribute-based and value-based clustering. An evaluation. In Proceedings of the 2004 conference on empirical methods in natural language processing, Barcelona, 158–165.
Amoia, Marilisa, and Claire Gardent. 2008. A test suite for inference involving adjectives. In Proceedings of the 6th international conference on language resources and evaluation, Marrakech, 631–637.
Baayen, R.H., R. Piepenbrock, and L. Gulikers. 1996. CELEX2. Philadelphia: Linguistic Data Consortium.
Baroni, Marco, Silvia Bernardini, Adriano Ferraresi, and Eros Zanchetta. 2009. The wacky wide web: A collection of very large linguistically processed web-crawled corpora. Journal of Language Resources and Evaluation 43(3): 209–226.
Barsalou, Lawrence W. 1992. Frames, concepts and conceptual fields. In Frames, Fields and Contrasts, ed. A. Lehrer and E.F. Kittay, 21–74. Hillsdale: Erlbaum.
Batista, Gustavo, Ronaldo Prati, and Maria Carolina Monard. 2004. A study of the behavior of several methods for balancing machine learning training data. SIGKDD Explorations 6: 20–29.
Beesley, Kenneth R. 1982. Evaluative adjectives as one-place predicates in Montague grammar. Journal of Semantics 1(3): 195–249.
Boleda, Gemma. 2006. Automatic acquisition of semantic classes for adjectives. Ph.D. dissertation, Pompeu Fabra University, Barcelona.
Buitelaar, Paul, Philipp Cimiano, and Bernardo Magnini. 2005. Ontology learning from text. An overview. In Ontology learning from text. Methods, evaluation and applications, ed. Paul Buitelaar, Philipp Cimiano, and Bernardo Magnini, 3–12. IOS Press: Amsterdam.
Cimiano, Philipp. 2006. Ontology learning and population from text. Algorithms, evaluation and applications. New York/London: Springer.
Ciravegna, F. 2000. Challenges in information extraction from texts for knowledge management. IEEE Intelligent Systems 16(6): 84–86.
Fellbaum, Christiane, ed. 1998. WordNet: An electronic lexical database. Cambridge: MIT.
Fleiss, Joseph L. 1971. Measuring nominal scale agreement among many raters. Psychological Bulletin 76(5): 378–382.
Guarino, Nicola. 1992. Concepts, attributes and arbitrary relations. Data & Knowledge Engineering 8: 249–261.
Hartung, Matthias, and Anette Frank. 2010a. A semi-supervised type-based classification of adjectives. Distinguishing properties and relations. In Proceedings of the 7th international conference on language resources and evaluation, Valletta, 1029–1036.
Hartung, Matthias, and Anette Frank. 2010b. A structured vector space model for hidden attribute meaning in adjective-noun phrases. In Proceedings of the 23rd international conference on computational linguistics (COLING), Beijing, 430–438.
Hartung, Matthias, and Anette Frank. 2011. Exploring supervised LDA models for assigning attributes to adjective-noun phrases. In Proceedings of the 2011 conference on empirical methods in natural language processing, Edinburgh, 540–551.
Hatzivassiloglou, Vasileios, and Kathleen R. McKeown. 1993. Towards the automatic identification of adjectival scales: Clustering adjectives according to meaning. In Proceedings of the 31st annual meeting of the association for computational linguistics, Columbus, 172–182.
Kamp, Hans. 1975. Two theories about adjectives. In Formal semantics of natural language, ed. E.L. Keenan, 123–155. Cambridge: Cambridge University Press.
Lapata, Mirella. 2001. The acquisition and modeling of lexical knowledge. A corpus-based investigation of systematic polysemy. Ph.D. dissertation, University of Edinburgh.
Levin, Beth. 1993. English verb classes and alternations. A preliminary investigation. Chicago: University of Chicago Press.
Levinson, Stephen C. 1983. Pragmatics. Cambridge: Cambridge University Press.
Marcus, Mitchell P., Mary Ann Marcinkiewicz, and Beatrice Santorini. 1993. Building a large annotated corpus of English. The Penn Treebank. Computational Linguistics 19(2): 313–330.
McNemar, Quinn. 1947. Note on the sampling error of the difference between correlated proportions or percentages. Psychometrika 12(2): 153–157.
Miller, Katherine J. 1998. Modifiers in WordNet. In WordNet. An electronic lexical database, ed. Christiane Fellbaum, 47–67. Cambridge: MIT.
Miyao, Yusuke, and Jun’ichi Tsujii. 2009. Supervised learning of a probabilistic lexicon of verb semantic classes. In Proceedings of the 2009 conference on empirical methods in natural language processing, Singapore, 1328–1337.
Montague, Richard. 1974. English as a formal language. In Formal philosophy, ed. R.H. Thomason, 247–270. New Haven: Yale University Press.
Pinto, H., and J. Martins. 2004. Ontologies. How can they be built? Knowledge and Information Systems 6(4): 441–464.
Raskin, Victor, and Sergei Nirenburg. 1998. An applied ontological semantic microtheory of adjective meaning for natural language processing. Machine Translation 13: 135–227.
Witten, Ian H., and Eibe Frank. 2005. Data mining. Practical machine learning tools and techniques, 2nd ed. San Francisco: Morgan Kaufmann.
Acknowledgements
The authors thank the annotators for their efforts and three anonymous reviewers of the final and an earlier version of this paper for highly valuable comments. As this article is a revised and extended version of Hartung and Frank (2010a), we also gratefully acknowledge the permission of the European Language Resources Association (ELRA) to republish the material.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Hartung, M., Frank, A. (2014). Distinguishing Properties and Relations in the Denotation of Adjectives: An Empirical Investigation. In: Gamerschlag, T., Gerland, D., Osswald, R., Petersen, W. (eds) Frames and Concept Types. Studies in Linguistics and Philosophy, vol 94. Springer, Cham. https://doi.org/10.1007/978-3-319-01541-5_8
Download citation
DOI: https://doi.org/10.1007/978-3-319-01541-5_8
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-01540-8
Online ISBN: 978-3-319-01541-5
eBook Packages: Humanities, Social Sciences and LawSocial Sciences (R0)