Abstract
This paper focuses on the research on how different agglomerative hierarchical clustering methods can be utilised to extract basic-level categories. Assuming three classical basic-levelness measures, namely category attentional slip, category utility and feature possession, and two hybrid measures, namely cue validity with global threshold and feature-possession, a multidendrogram approach is studied. In particular, different proximity measures and linkage criteria are thoroughly investigated against three datasets representing different characteristics of typical data in cyber-physical systems. Performed investigation highlights how different clustering settings affect basic-levelness measures and indicates that additional pruning of features is required.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
Whatbird - https://www.whatbird.com/; Mushroom dataset - https://www.openml.org/d/24; Zoo dataset - https://www.openml.org/d/62.
- 2.
For CAS measure we assume that attention randomly slips with probability \(p = 0.5\) (analogous to [5]), while for (CVGT) global threshold is equal to 0.7.
- 3.
In case of CAS measure we utilise it’s complement.
References
Bornstein, M.H., Arterberry, M.E.: The development of object categorization in young children: hierarchical inclusiveness, age, perceptual attribute, and group versus individual analyses. Dev. Psychol. 46(2), 350–365 (2010)
Corter, J.E., Gluck, M.A.: Explaining basic categories: feature predictability and information. Psychol. Bullet. 111(2), 291–303 (1992)
Everitt, B., Landau, S., Leese, M., Stahl, D.: Cluster Analysis Wiley Series in Probability and Statistics. Wiley, New Jersey (2011)
Fernández, A., Gómez, S.: Solving non-uniqueness in agglomerative hierarchical clustering using multidendrograms. J. Classif. 25(1), 43–65 (2008)
Gosselin, F., Schyns, P.: Debunking the Basic Level. In: Cognitive Science Society (US) Conference/Proceedings, pp. 277–282 (1997)
Harnad, S.: To Cognize is to Categorize: Cognition is Categorization, Handbook of Categorization in Cognitive Science (Second Edition), Elsevier, 21–54 (2017)
Jones, G.V.: Identifying basic categories. Psych. Bullet. 94(3), 423–428 (1983)
Mulka, M., Lorkiewicz, W.: Measures for extracting basic-level categories. In: Proceedings CISP-BMEI, pp. 1–6. IEEE (2018)
Mulka, M., Lorkiewicz, W., Katarzyniak, R.P.: Extraction of basic-level categories using dendrogram and multidendrogram. In: Proceedings ICNC-FSKD, Springer (2019). (In press)
Rosch, E.: Basic objects in natural categories. Working paper. Language Behavior Research Laboratory, University of California (1975)
Saxena, A., Prasad, M., Gupta, A., Bharill, N., Patel, O.P., Tiwari, A., Er, M.J., Ding, W., Lin, C.T.: A review of clustering techniques and developments. Neurocomputing 267, 664–681 (2017)
Acknowledgment
This research was carried out at Wrocław University of Science and Technology (Poland) under Grant 0401/0190/18 titled Models and Methods of Semantic Communication in Cyber-Physical Systems.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Mulka, M., Lorkiewicz, W. (2020). Different Hierarchical Clustering Methods in Basic-Level Extraction Using Multidendrogram. In: Świątek, J., Borzemski, L., Wilimowska, Z. (eds) Information Systems Architecture and Technology: Proceedings of 40th Anniversary International Conference on Information Systems Architecture and Technology – ISAT 2019. ISAT 2019. Advances in Intelligent Systems and Computing, vol 1051. Springer, Cham. https://doi.org/10.1007/978-3-030-30604-5_18
Download citation
DOI: https://doi.org/10.1007/978-3-030-30604-5_18
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-30603-8
Online ISBN: 978-3-030-30604-5
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)