Abstract
In this paper we introduce the ttcs system, so named after Terms To Conceptual Spaces, that exploits a resource-driven approach relying on BabelNet, NASARI and ConceptNet. ttcs takes in input a term and its context of usage and produces as output a specific type of vector-based semantic representation, where conceptual information is encoded through the Conceptual Spaces (a geometric framework for common-sense knowledge representation and reasoning). The system has been evaluated in a twofold experimentation. In the first case we assessed the quality of the extracted common-sense conceptual information with respect to human judgments with an online questionnaire. In the second one we compared the performances of a conceptual categorization system that was run twice, once fed with extracted annotations and once with hand-crafted annotations. In both cases the results are encouraging and provide precious insights to make substantial improvements.
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
Similar content being viewed by others
Notes
- 1.
Recently the convexity constraint of conceptual spaces has been argued as a plausible but not necessary condition for the characterisation of concepts within this framework in the case, for example, of the adoption of non-euclidean metrics (see [4]). In our case-study we considered such constraint as proposed in the original theory since we didn’t consider non-euclidean metrics.
- 2.
In fact, we remark that differently from CSs, formal ontologies are not suited for representing defeasible, prototypical knowledge and for dealing with the corresponding typicality-based conceptual reasoning (e.g., non-monotonic inference). For example, for the concept dog, OpenCyc does not represent that “typically” dogs bark and woof because common-sense traits are not necessary/sufficient for defining this category.
- 3.
Resources marked with emphasized fonts are harmonized in UBY in both the English and the German version.
- 4.
Typically, the context is composed by one or more sentences; without loss of generality, in the present setting the context has been retrieved by accessing the DBPedia page associated to t.
- 5.
NASARI unified vectors are composed by a head concept (represented by its ID in the first position) and a body, that is a list of synsets related to the head concept. Each synset ID is followed by a number that grasps its correlation with the head concept. It is worth noting that in order to reduce the number of required accesses to BabelNet we built an all-in-one resource that maps each ID referred in NASARI vectors onto its synset terms.
- 6.
\(\alpha \) is presently set to 100.
- 7.
\(\beta \) is presently set to 3.
- 8.
We note that the presence of \(t'_{i}\) in the vector of c is guaranteed only if \(t'_{i}\) was detected as relevant through the first relevance condition. So, if \(t'_{i}\) does not appear in the vector of \(c^t\), the identification process fails, and the term will not be added to the result set C.
- 9.
Correctly identified concepts are those for which the whole procedure produces an output.
- 10.
Questionnaires are available at: http://goo.gl/am0S2f.
- 11.
- 12.
As the ratio between the number of deletions expressed for a given statement and the number of assessments obtained by that statement: e.g., the statement ‘Soap has function of scent’ has been questioned by 2 participants out of 12. The agreement on such deletion was computed as \(2/12=16.7\,\%\).
- 13.
In essence, the employed system executes a two-steps categorization process: it first computes a result based on Conceptual Spaces, and it then checks the validity of the obtained result against an ontological knowledge base.
References
Navigli, R., Ponzetto, S.P.: BabelNet: the automatic construction, evaluation and application of a wide-coverage multilingual semantic network. Artif. Intell. 193, 217–250 (2012)
Camacho-Collados, J., Pilehvar, M.T., Navigli, R.: NASARI: a novel approach to a semantically-aware representation of items. In: Proceedings of NAACL, pp. 567–577 (2015)
Speer, R., Havasi, C.: Representing general relational knowledge in ConceptNet 5. In: LREC, pp. 3679–3686 (2012)
Hernández-Conde, J.V.: A case against convexity in conceptual spaces. Synthese, 1–27 (2016)
Gärdenfors, P.: The Geometry of Meaning: Semantics Based on Conceptual Spaces. MIT Press, Cambridge (2014)
Rosch, E.: Cognitive representations of semantic categories. J. Exp. Psychol. Gen. 104, 192–233 (1975)
Lenat, D.B., Prakash, M., Shepherd, M.: CYC: using common sense knowledge to overcome brittleness and knowledge acquisition bottlenecks. AI Mag. 6, 65 (1985)
Frixione, M., Lieto, A.: Representing concepts in formal ontologies: compositionality vs. typicality effects. Logic Logical Philos. 21, 391–414 (2012)
Reeve, L., Han, H.: Survey of semantic annotation platforms. In: Proceedings of the 2005 ACM symposium on Applied computing, pp. 1634–1638. ACM (2005)
Derrac, J., Schockaert, S.: Inducing semantic relations from conceptual spaces: a data-driven approach to plausible reasoning. Artif. Intell. 228, 66–94 (2015)
Pennington, J., Socher, R., Manning, C.D.: Glove: global vectors for word representation. EMNLP 14, 1532–1543 (2014)
Mikolov, T., Sutskever, I., Chen, K., Corrado, G.S., Dean, J.: Distributed representations of words and phrases and their compositionality. In: Advances in Neural Information Processing Systems, pp. 3111–3119 (2013)
Turney, P.D., Pantel, P., et al.: From frequency to meaning: vector space models of semantics. J. Artif. Intell. Res. 37(1), 141–188 (2010)
Miller, G.A.: WordNet: a lexical database for English. Commun. ACM 38(11), 39–41 (1995)
Baker, C.F., Fillmore, C.J., Lowe, J.B.: The Berkeley framenet project. In: Proceedings of the 17th International Conference on Computational Linguistics. Association for Computational Linguistics, vol. 1, pp. 86–90 (1998)
Levin, B.: English Verb Classes and Alternations: A Preliminary Investigation. University of Chicago Press, Chicago (1993)
Havasi, C., Speer, R., Alonso, J.: ConceptNet: a lexical resource for common sense knowledge. Recent Adv. Nat. Lang. Process. V Sel. Pap. RANLP 309, 269 (2007)
Eckle-Kohler, J., Gurevych, I., Hartmann, S., Matuschek, M., Meyer, C.M.: UBY-LMF-a uniform model for standardizing heterogeneous lexical-semantic resources in ISO-LMF. In: LREC, pp. 275–282 (2012)
Gurevych, I., Eckle-Kohler, J., Hartmann, S., Matuschek, M., Meyer, C.M., Wirth, C.: UBY: a large-scale unified lexical-semantic resource based on LMF. In: Proceedings of the 13th Conference of the European Chapter of the Association for Computational Linguistics. Association for Computational Linguistics, pp. 580–590 (2012)
McCrae, J., Montiel-Ponsoda, E., Cimiano, P.: Integrating WordNet and Wiktionary with lemon. In: Chiarcos, C., Nordhoff, S., Hellmann, S. (eds.) Linked Data in Linguistics, pp. 25–34. Springer, Heidelberg (2012)
Buitelaar, P., Cimiano, P., Haase, P., Sintek, M.: Towards linguistically grounded ontologies. In: Aroyo, L., et al. (eds.) ESWC 2009. LNCS, vol. 5554, pp. 111–125. Springer, Heidelberg (2009). doi:10.1007/978-3-642-02121-3_12
Tonelli, S., Pianta, E.: A novel approach to mapping framenet lexical units to Wordnet synsets. In: Proceedings of the Eighth International Conference on Computational Semantics. Association for Computational Linguistics, pp. 342–345 (2009)
Ferrández, O., Ellsworth, M., Munoz, R., Baker, C.F.: Aligning FrameNet and WordNet based on semantic neighborhoods. LREC 10, 310–314 (2010)
Pilehvar, M.T., Jurgens, D., Navigli, R.: Align, disambiguate and walk: a unified approach for measuring semantic similarity. In: ACL, vol. 1, pp. 1341–1351 (2013)
Bejan, A., Marden, J.H.: Constructing animal locomotion from new thermadynamics theory. Am. Sci. 94, 342–349 (2006)
Tonelli, S., Pighin, D.: New features for framenet: Wordnet mapping. In: Proceedings of the Thirteenth Conference on Computational Natural Language Learning. Association for Computational Linguistics, pp. 219–227 (2009)
De Cao, D., Croce, D., Basili, R.: Extensive evaluation of a framenet-wordnet mapping resource. In: LREC (2010)
Lieto, A., Minieri, A., Piana, A., Radicioni, D.P.: A knowledge-based system for prototypical reasoning. Connection Sci. 27, 137–152 (2015)
Lieto, A., Radicioni, D.P., Rho, V.: A common-sense conceptual categorization system integrating heterogeneous proxytypes and the dual process of reasoning. In: Proceedings of the International Joint Conference on Artificial Intelligence (IJCAI), pp. 875–881. AAAI Press, Buenos Aires (2015)
Lieto, A., Radicioni, D.P., Rho, V.: Dual PECCS: a cognitive system for conceptual representation and categorization. J. Exp. Theor. Artif. Intell., 1–20 (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing AG
About this paper
Cite this paper
Lieto, A., Mensa, E., Radicioni, D.P. (2016). A Resource-Driven Approach for Anchoring Linguistic Resources to Conceptual Spaces. In: Adorni, G., Cagnoni, S., Gori, M., Maratea, M. (eds) AI*IA 2016 Advances in Artificial Intelligence. AI*IA 2016. Lecture Notes in Computer Science(), vol 10037. Springer, Cham. https://doi.org/10.1007/978-3-319-49130-1_32
Download citation
DOI: https://doi.org/10.1007/978-3-319-49130-1_32
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-49129-5
Online ISBN: 978-3-319-49130-1
eBook Packages: Computer ScienceComputer Science (R0)