Conceptual clustering in a first order logic representation
ECAI '92 Proceedings of the 10th European conference on Artificial intelligence
Inductive learning of characteristic concept descriptions from small sets of classified examples
ECML-94 Proceedings of the European conference on machine learning on Machine Learning
Theories for mutagenicity: a study in first-order and feature-based induction
Artificial Intelligence - Special volume on empirical methods
Relational instance-based learning with lists and terms
Machine Learning - Special issue on inducive logic programming
Extensions to the k-Means Algorithm for Clustering Large Data Sets with Categorical Values
Data Mining and Knowledge Discovery
ECML '93 Proceedings of the European Conference on Machine Learning
ECML '97 Proceedings of the 9th European Conference on Machine Learning
Top-Down Induction of Clustering Trees
ICML '98 Proceedings of the Fifteenth International Conference on Machine Learning
K-means Clustering Algorithm for Categorical Attributes
DaWaK '99 Proceedings of the First International Conference on Data Warehousing and Knowledge Discovery
Distance Induction in First Order Logic
ILP '97 Proceedings of the 7th International Workshop on Inductive Logic Programming
Distance Between Herbrand Interpretations: A Measure for Approximations to a Target Concept
ILP '97 Proceedings of the 7th International Workshop on Inductive Logic Programming
A Framework for Defining Distances Between First-Order Logic Objects
ILP '98 Proceedings of the 8th International Workshop on Inductive Logic Programming
Term Comparisons in First-Order Similarity Measures
ILP '98 Proceedings of the 8th International Workshop on Inductive Logic Programming
Relational Distance-Based Clustering
ILP '98 Proceedings of the 8th International Workshop on Inductive Logic Programming
Querying and Merging Heterogeneous Data by Approximate Joins on Higher-Order Terms
ILP '08 Proceedings of the 18th international conference on Inductive Logic Programming
K-means based approaches to clustering nodes in annotated graphs
ISMIS'11 Proceedings of the 19th international conference on Foundations of intelligent systems
| Hi-index | 0.00 |
In this paper, we present an in-depth evaluation of two approaches of extending k-means clustering to work on first-order representations. The first-approach, k-medoids, selects its cluster center from the given set of instances, and is thus limited in its choice of centers. The second approach, k-prototypes, uses a heuristic prototype construction algorithm that is capable of generating new centers. The two approaches are empirically evaluated on a standard benchmark problem with respect to clustering quality and convergence. Results show that in this case indeed the k-medoids approach is a viable and fast alternative to existing agglomerative or top-down clustering approaches even for a small-scale dataset, while k-prototypes exhibited a number of deficiencies.