Evaluation of Topographic Clustering and Its Kernelization
نویسندگان
چکیده
We consider the topographic clustering task and focus on the problem of its evaluation, which enables to perform model selection: topographic clustering algorithms, from the original Self Organizing Map to its extension based on kernel (STMK), can be viewed in the unified framework of constrained clustering. Exploiting this point of view, we discuss existing quality measures and we propose a new criterion based on an F-measure, which combines a compacity with an organization criteria and extend it to their kernel-based version.
منابع مشابه
The K-discs Algorithm and Its Kernelization
A new clustering algorithm called k-discs is introduced. This algorithm, though similar to k-means, addresses the de ciencies of k-means as well as its variant k-subspaces. The k-discs algorithm is applied to the recovery of manifolds from noisy samplings. A kernelization of k-discs is exhibited, and the advantages of this modi cation are demonstrated on synthetic data. For this project, the ke...
متن کاملA Document of Ant Clustering By the Result of Topographic Mapping Intention
The clustering and topographic mapping is a motivated model for explaining two types of developing behavior observed in real ant colonies. Existing work demonstrated some promising characteristics of the heuristic but did not extend to a rigorous investigation of its capabilities. We are using an Improved Technique called ATTA includes adaptive, heterogeneous ants, a time-dependent transporting...
متن کاملCore Decomposition in Directed Networks: Kernelization and Strong Connectivity
In this paper, we propose a method allowing decomposition of directed networks into cores, which final objective is the detection of communities. We based our approach on the fact that a community should be composed of elements having communication in both directions. Therefore, we propose a method based on digraph kernelization and strongly p-connected components. By identifying cores, one can...
متن کاملAnt-Based Clustering and Topographic Mapping
Ant-based clustering and sorting is a nature-inspired heuristic first introduced as a model for explaining two types of emergent behavior observed in real ant colonies. More recently, it has been applied in a data-mining context to perform both clustering and topographic mapping. Early work demonstrated some promising characteristics of the heuristic but did not extend to a rigorous investigati...
متن کاملAverage Parameterization and Partial Kernelization for Computing Medians
We propose an effective polynomial-time preprocessing strategy for intractable median problems. Developing a new methodological framework, we show that if the input instances of generally intractable problems exhibit a sufficiently high degree of similarity between each other on average, then there are efficient exact solving algorithms. In other words, we show that the median problems Swap Med...
متن کامل