Cookies Policy
We use cookies to improve our site and your experience. By continuing to browse our site you accept our cookie policy. Find out More
Close
  • Menu
Interest
Topics
Details

Details

  • Name

    Pedro Manuel Ribeiro
  • Cluster

    Computer Science
  • Role

    Senior Researcher
  • Since

    03rd May 2010
Publications

2017

Evolutionary role mining in complex networks by ensemble clustering

Authors
Choobdar, Sarvenaz; Ribeiro, PedroManuelPinto; Silva, FernandoM.A.;

Publication
Proceedings of the Symposium on Applied Computing, SAC 2017, Marrakech, Morocco, April 3-7, 2017

Abstract

2017

Extending the Applicability of Graphlets to Directed Networks

Authors
Aparicio, D; Ribeiro, P; Silva, F;

Publication
IEEE-ACM TRANSACTIONS ON COMPUTATIONAL BIOLOGY AND BIOINFORMATICS

Abstract
With recent advances in high-throughput cell biology, the amount of cellular biological data has grown drastically. Such data is often modeled as graphs (also called networks) and studying them can lead to new insights intomolecule-level organization. A possible way to understand their structure is by analyzing the smaller components that constitute them, namely network motifs and graphlets. Graphlets are particularly well suited to compare networks and to assess their level of similarity due to the rich topological information that they offer but are almost always used as small undirected graphs of up to five nodes, thus limiting their applicability in directed networks. However, a large set of interesting biological networks such asmetabolic, cell signaling, or transcriptional regulatory networks are intrinsically directional, and using metrics that ignore edge direction may gravely hinder information extraction. Our main purpose in this work is to extend the applicability of graphlets to directed networks by considering their edge direction, thus providing a powerful basis for the analysis of directed biological networks. We tested our approach on two network sets, one composed of synthetic graphs and another of real directed biological networks, and verified that they were more accurately grouped using directed graphlets than undirected graphlets. It is also evident that directed graphlets offer substantially more topological information than simple graph metrics such as degree distribution or reciprocity. However, enumerating graphlets in large networks is a computationally demanding task. Our implementation addresses this concern by using a state-of-the-art data structure, the g-trie, which is able to greatly reduce the necessary computation. We compared our tool to other state-of-the art methods and verified that it is the fastest general tool for graphlet counting.

2017

TensorCast: Forecasting with Context Using Coupled Tensors (Best Paper Award)

Authors
de Araujo, MR; Pinto Ribeiro, PM; Faloutsos, C;

Publication
2017 IEEE International Conference on Data Mining, ICDM 2017, New Orleans, LA, USA, November 18-21, 2017

Abstract

2017

Scalable subgraph counting using MapReduce

Authors
Eddin, AN; Pinto Ribeiro, PM;

Publication
Proceedings of the Symposium on Applied Computing, SAC 2017, Marrakech, Morocco, April 3-7, 2017

Abstract
Networks are powerful in representing a wide variety of systems in many fields of study. Networks are composed of smaller substructures (subgraphs) that characterize them and give important information related to their topology and functionality. Therefore, discovering and counting these subgraph patterns is very important towards mining the features of networks. Algorithmically, subgraph counting in a network is a computationally hard problem and the needed execution time grows exponentially as the size of the subgraph or the network increases. The main goal of this paper is to contribute towards subgraph search, by providing an accessible and scalable parallel methodology for counting subgraphs. For that we present a dynamic iterative MapReduce strategy to parallelize algorithms that induce an unbalanced search tree, and apply it in the subgraph counting realm. At the core of our methods lies the g-trie, a state-of-the-art data structure that was created precisely for this task. Our strategy employs an adaptive time threshold and an efficient work-sharing mechanism to dynamically do load balancing between the workers. We evaluate our implementations using Spark on a large set of representative complex networks from different fields. The results obtained are very promising and we achieved a consistent and almost linear speedup up to 32 cores, with an average efficiency close to 80%. To the best of our knowledge this is the fastest and most scalable method for subgraph counting within the MapReduce programming model. Copyright 2017 ACM.

2016

FastStep: Scalable Boolean Matrix Decomposition

Authors
Araujo, M; Ribeiro, P; Faloutsos, C;

Publication
ADVANCES IN KNOWLEDGE DISCOVERY AND DATA MINING, PAKDD 2016, PT I

Abstract
Matrix Decomposition methods are applied to a wide range of tasks, such as data denoising, dimensionality reduction, co-clustering and community detection. However, in the presence of boolean inputs, common methods either do not scale or do not provide a boolean reconstruction, which results in high reconstruction error and low interpretability of the decomposition. We propose a novel step decomposition of boolean matrices in non-negative factors with boolean reconstruction. By formulating the problem using threshold operators and through suitable relaxation of this problem, we provide a scalable algorithm that can be applied to boolean matrices with millions of non-zero entries. We show that our method achieves significantly lower reconstruction error when compared to standard state of the art algorithms. We also show that the decomposition keeps its interpretability by analyzing communities in a flights dataset (where the matrix is interpreted as a graph in which nodes are airports) and in a movie-ratings dataset with 10 million non-zeros.

Supervised
thesis

2016

Communities and Anomaly Detection in Large Edge-Labeled Graphs

Author
Miguel Ramos de Araújo

Institution
UP-FCUP

2016

Pattern Discovery in Complex Networks

Author
David Oliveira Aparício

Institution
UP-FCUP

2016

Large Scale Parallel Subgraph Search

Author
Ahmad Naser Eddin

Institution
UP-FCUP

2016

Bomberman as an Artificial Intelligence Platform

Author
Manuel António da Cruz Lopes

Institution
UP-FCUP

2015

On the Characterization and Comparison of Complex Networks

Author
Sarvenaz Choobdar

Institution
UP-FCUP