László Aszalós
YOU?
Author Swipe
View article: Decompose Boolean Matrices with Correlation Clustering
Decompose Boolean Matrices with Correlation Clustering Open
One of the tasks of data science is the decomposition of large matrices in order to understand their structures. A special case of this is when we decompose relations, i.e., logical matrices. In this paper, we present a method based on the…
View article: Graph approximation on similarity based rough sets
Graph approximation on similarity based rough sets Open
Correlation clustering is a widely used technique in data mining. The clusters contain objects, which are typically similar to one another and different from objects from other groups. In the authors previous works the possible usage of co…
View article: Selecting representatives
Selecting representatives Open
We use representatives to reduce complexity in many areas of life.Clusters are often replaced with their centre, and then these representatives are used to classify new objects.If the objects are described as a vector of real numbers, then…
View article: Correlation Clustering: Let All The Flowers Bloom!
Correlation Clustering: Let All The Flowers Bloom! Open
Correlation clustering is a NP-hard problem, and for large signed graphs finding even just a good approximation of the optimal solution is a hard task.In this article we examine the effect of ranking the nodes and processing them in order …
View article: Different Types of Search Algorithms for Rough Sets
Different Types of Search Algorithms for Rough Sets Open
Based on the available information in many cases it can happen that two objects cannot be distinguished. If a set of data is given and in this settwo objects have the same attribute values, then these two objects are called indiscernible. …
View article: Finding the representative in a cluster using correlation clustering
Finding the representative in a cluster using correlation clustering Open
Correlation clustering is a widely used technique in data mining. The clusters contain objects, which are typically similar to each other and different from objects from other groups. It can be an interesting task to find the member, which…
View article: Visualization of tolerance relations
Visualization of tolerance relations Open
Previously we generated tolerance relations from the distance of objects.In this paper the situation is reversed: from any tolerance relation we generate a 2D representation in a such a way that the nodes of similar objects are close, whil…
View article: Correlation clustering: a parallel approach?
Correlation clustering: a parallel approach? Open
Correlation clustering is a NP-hard problem, and for large graphs finding even just a good approximation of the optimal solution is a hard task.In previous articles we have suggested a contraction method and its divide and conquer variant.…
View article: Correlation clustering of graphs and integers
Correlation clustering of graphs and integers Open
Correlation clustering can be modeled in the following way. Let A be a nonempty set, and ∼ be a symmetric binary relation on A. Consider a partition (clustering) P of A. We say that two distinct elements a, b ∈ A are in conflict, if a ∼ b,…
View article: Correlation clustering: divide and conquer
Correlation clustering: divide and conquer Open
The correlation clustering is an NP-hard problem, hence its solving methods do not scale well.The contraction method and its improvement enable us to construct a divide and conquer algorithm, which could help us to clustering bigger sets.I…
View article: Correlation clustering by contraction
Correlation clustering by contraction Open
We suggest an effective method for solving the problem of correlation clustering. This method is based on an extension of a partial tolerance relation to clusters. We present several implementation of this method using different data struc…
View article: Email labelling by rough clustering
Email labelling by rough clustering Open
Previously, there were little possibilities to sort mails and later emails: we could only arrange them into folders.One mail or email could be put into exactly one folder, based on sender, subject or priority.Later in Gmail the labelling o…
View article: Rough Classification in Incomplete Databases by Correlation Clustering
Rough Classification in Incomplete Databases by Correlation Clustering Open
%&'() * +,-./-,0 1&* 2 '3 45/2 670 3 2 * +,-8790 7'7/4:(/;&0 + <=>?@AB? ./* C7',