José E. Medina-Pagola

Frequent approximate subgraphs as features for graph-based image classification

The use of approximate graph matching for frequent subgraph mining has been identified in different applications as a need. To meet this need, several algorithms have been developed, but there are applications where it has not been used yet, for example image classification. In this paper, a new algorithm for mining frequent connected subgraphs over undirected and labeled graph collections VEAM (Vertex and Edge Approximate graph Miner) is presented. Slight variations of the data, keeping the topology of the graphs, are allowed in this algorithm. Approximate matching in existing algorithm (APGM) is only performed on vertex label set. In VEAM, the approximate matching between edge label set in frequent subgraph mining is included in the mining process. Also, a framework for graph-based image classification is introduced. The approximate method of VEAM was tested on an artificial image collection using a graph-based image representation proposed in this paper. The experimentation on this collection shows that our proposal gets better results than graph-based image classification using some algorithms reported in related work.

OClustR: A new graph-based algorithm for overlapping clustering

Clustering is a Data Mining technique, which has been widely used in many practical applications. From these applications, there are some, like social network analysis, topic detection and tracking, information retrieval, categorization of digital libraries, among others, where objects may belong to more than one cluster; however, most clustering algorithms build disjoint clusters. In this work, we introduce OClustR, a new graph-based clustering algorithm for building overlapping clusters. The proposed algorithm introduces a new graph-covering strategy and a new filtering strategy, which together allow to build overlapping clusterings more accurately than those built by previous algorithms. The experimental evaluation, conducted over several standard collections, showed that our proposed algorithm builds less clusters than those built by the previous related algorithms. Additionally, OClustR builds clusters with overlapping closer to the real overlapping in the collections than the overlapping generated by other clustering algorithms.

A new proposal for graph-based image classification using frequent approximate subgraphs

Graph-based data representations are an important research topic due to the suitability of this kind of data structure to model entities and the complex relations among them. In computer vision, graphs have been used to model images in order to add some high level information (relations) to the low-level representation of individual parts. How to deal with these representations for specific tasks is not easy due to the complexity of the data structure itself. In this paper we propose to use a graph mining technique for image classification, introducing approximate patterns discovery in the mining process in order to allow certain distortions in the data being modeled. We are proposing to combine a powerful graph-based image representation adapted to this specific task and frequent approximate subgraph (FAS) mining algorithms in order to classify images. In the case of image representation we are proposing to use more robust descriptors than our previous approach in this topic, and we also suggest a criterion to select the isomorphism threshold for the graph mining step. This proposal is tested in two well-known collections to show the improvement with respect to the previous related works. HighlightsWe propose a new framework for image classification, which uses frequent approximate subgraph patterns as features.We propose to compute automatically the substitution matrices needed in the process, instead of using expert knowledge.We propose to use a new graph-based image representation.We propose a criterion for selecting isomorphism threshold for the graph mining process.

An algorithm based on density and compactness for dynamic overlapping clustering

Most clustering algorithms organize a collection of objects into a set of disjoint clusters. Although this approach has been successfully applied in unsupervised learning, there are several applications where objects could belong to more than one cluster. Overlapping clustering is an alternative in those contexts like social network analysis, information retrieval and bioinformatics, among other problems where non-disjoint clusters appear. In addition, there are environments where the collection changes frequently and the clustering must be updated; however, most of the existing overlapping clustering algorithms are not able to efficiently update the clustering. In this paper, we introduce a new overlapping clustering algorithm, called DClustR, which is based on the graph theory approach and it introduces a new strategy for building more accurate overlapping clusters than those built by state-of-the-art algorithms. Moreover, our algorithm introduces a new strategy for efficiently updating the clustering when the collection changes. The experimentation conducted over several standard collections shows the good performance of the proposed algorithm, wrt. accuracy and efficiency.

Full duplicate candidate pruning for frequent connected subgraph mining

Support calculation and duplicate detection are the most challenging and unavoidable subtasks in frequent connected subgraph (FCS) mining. The most successful FCS mining algorithms have focused on optimizing these subtasks since the existing solutions for both subtasks have high computational complexity. In this paper, we propose two novel properties that allow removing all duplicate candidates before support calculation. Besides, we introduce a fast support calculation strategy based on embedding structures. Both properties and the new embedding structure are used for designing two new algorithms: gdFil for mining all FCSs; and gdClosed for mining all closed FCSs. The experimental results show that our proposed algorithms get the best performance in comparison with other well known algorithms.

A dynamic clustering algorithm for building overlapping clusters

Clustering is a Data Mining technique which has been widely used in many practical applications. In some of these applications like, medical diagnosis, categorization of digital libraries, topic detection and others, the objects could belong to more than one cluster. However, most of the clustering algorithms generate disjoint clusters. Moreover, processing additions, deletions and modifications of objects in the clustering built so far, without having to rebuild the clustering from the beginning is an issue that has been little studied. In this paper, we introduce DCS, a clustering algorithm which includes a new graph-cover strategy for building a set of clusters that could overlap, and a strategy for dynamically updating the clustering, managing multiple additions and/or deletions of objects. The experimental evaluation conducted over different collections demonstrates the good performance of the proposed algorithm.

On Speeding up Frequent Approximate Subgraph Mining

Frequent approximate subgraph (FAS) mining has become an interesting task with wide applications in several domains of science. Most of the previous studies have been focused on reducing the search space or the number of canonical form (CF) tests. CF-tests are commonly used for duplicate detection; however, these tests affect the efficiency of mining process because they have high computational complexity. In this paper, two prunes are proposed, which allow decreasing the label space, the number of candidates and the number of CF-tests. The proposed prunes are already used and validated in two reported FAS miners by speeding up their mining processes in artificial graph collections.

Extension of Canonical Adjacency Matrices for Frequent Approximate Subgraph Mining on Multi-Graph Collections

Into the data mining field, frequent approximate subgraph (FAS) mining has become an important technique with a broad spectrum of real-life applications. This fact is because several real-life phenomena can be modeled by graphs. In the literature, several algorithms have been reported for mining frequent approximate patterns on simple-graph collections; however, there are applications where more complex data structures, as multi-graphs, are needed for modeling the problem. But to the best of our knowledge, there is no FAS mining algorithm designed for dealing with multi-graphs. Therefore, in this paper, a canonical form (CF) for simple-graphs is extended to allow representing multi-graphs and a state-of-the-art algorithm for FAS mining is also extended for processing multi-graph collections by using the extended CF. Our experiments over different synthetic and real-world multi-graph collections show that the proposed algorithm has a good performance in terms of runtime and scalability. Additionally, we show the usefulness of the patterns computed by our algorithm in an image classification problem where images are represented as multi-graphs.

A New Method Based on Graph Transformation for FAS Mining in Multi-graph Collections

Currently, there has been an increase in the use of frequent approximate subgraph FAS mining for different applications like graph classification. In graph classification tasks, FAS mining algorithms over graph collections have achieved good results, specially those algorithms that allow distortions between labels, keeping the graph topology. However, there are some applications where multi-graphs are used for data representation, but FAS miners have been designed to work only with simple-graphs. Therefore, in this paper, in order to deal with multi-graph structures, we propose a method based on graph transformations for FAS mining in multi-graph collections.

A new algorithm for mining frequent connected subgraphs based on adjacency matrices

Most of the Frequent Connected Subgraph Mining (FCSM) algorithms have been focused on detecting duplicate candidates using canonical form (CF) tests. CF tests have high computational complexity, which affects the efficiency of graph miners. In this paper, we introduce novel properties of the canonical adjacency matrices for reducing the number of CF tests in FCSM. Based on these properties, a new algorithm for frequent connected subgraph mining called grCAM is proposed. The experiments on real world datasets show the impact of the proposed properties in FCSM. Besides, the performance of our algorithm is compared against some other reported algorithms.