Linha de Pesquisa: Inteligência Artificial

Resumo do trabalho:

Network based classification has demonstrated its value especially due to its intrinsic ability to capture networked data properties (e.g., structural and dynamical). However, its performance is highly dependent upon the network architecture. In this sense, we present a network architecture optimization method by employing genetic algorithms (GAs) for the classification via characterization of importance. The importance based classification is a recent network classification technique that uses the pagerank measure to capture the underlying data relationship. In particular, we hypothesize that the prominent characteristics of GAs, such as their robust search mechanism and binary representation, may provide a more effective network architecture. Further, in an effort to capture the relationships between the networked data, we also analyze, despite pagerank, other network measures, namely degree, betweenness, closeness, and shortest path length. In summary, the experimental findings using real data sets demonstrated that the proposed algorithm outperforms the widely used k-nearest neighbors graph method in terms of classification accuracy. It also shows competitive results against a state-of-the-art network optimization technique based on swarm intelligence. Meanwhile, for the network measures, the results revealed that pagerank and degree produced the best outcomes and statistically outperformed all other network measures in terms of predictive capability and robustness. Our technique was also applied to the detection of autism spectrum disorder from salivary data processed by the attenuated total reflectance Fourier-transform infrared (ATR-FTIR) spectroscopy. In the experiments, it was able to outperform linear discriminant analysis, a widely adopted technique in ATR-FTIR analysis, as well as support vector machine, a state-of-the art technique for such problems. Moreover, such results give evidences about the potential of our approach in dealing with such a difficult problem, characterized by high- dimensional data and arbitrary distributions.

Link da transmissão da defesa:

https://teams.microsoft.com/l/meetup-join/19%3ameeting_MzI4Zjc5ZDYtYjRhM...