Feature selection for optical network design via a new mutual information estimator

Depizzol, Daniela Bertolini; Montalvão, Jugurta; Lima, Fabio de Oliveira; Paiva, Marcia Helena Moreira; Segatto, Marcelo E. V.

doi:10.1016/j.eswa.2018.04.018

Cited by 9 publications

(9 citation statements)

References 26 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…No trabalho [Depizzol et al 2018] foram estudadas 315 invariantes e o resultado revelou um subconjunto de invariantes que possuem correlac ¸ão com o número de comprimentos de onda necessários em uma rede WDM. Neste trabalho vamos considerar invariantes de grafos relacionadas a esse subconjunto (Grau nodal, Centralidade de intermediac ¸ão do enlace e Centralidade de excentricidade) e algumas adicionais, e identificar a correlac ¸ão entre essas invariantes e o desempenho da rede, calculado a partir do coeficiente de bloqueio.…”

Section: Definic ¸ãO De Grafos E Invariantesunclassified

“…Atualmente são conhecidas mais de 300 invariantes de grafo, e os autores de [Depizzol et al 2018] apresentaram um subconjunto aplicável à otimizac ¸ão do número de comprimentos de onda em redes de telecomunicac ¸ões. Neste artigo, vamos buscar nesse subconjunto e em outros invariantes de grafos, aqueles que têm correlac ¸ão com o desempenho da rede, medido a partir do coeficiente de bloqueio, que representa a razão entre o número de requisic ¸ões bloqueadas pela rede e o número total de requisic ¸ões demandadas.…”

unclassified

See 1 more Smart Citation

Uma análise da relação entre invariantes de grafos e coeficiente de bloqueio em Redes Ópticas Elásticas

Wenzel¹,

Neris²,

Paiva³

et al. 2021

Anais Do XXXIX Simpósio Brasileiro De Redes De Computadores E Sistemas Distribuídos (SBRC 2021)

View full text Add to dashboard Cite

O caráter dinâmico do atual aperfeiçoamento de técnicas de engenharia de rede e tráfego para otimizar a capacidade de transmissão. Neste artigo, demonstramos a correlação entre o desempenho de topologias de rede, medido em termos de bloqueio de requisições, e invariantes de grafo. São apresentadas e quantiﬁcadas correlações que possibilitam conhecermos a sensibilidade de cada invariante no comportamento da rede. A invariante desvio padrão da centralidade de intermediação do enlace apresentou a mais forte correlação com o desempenho das topologias, e serve como uma referência para tomada de decisões em tarefas de dimensionamento e otimização de redes.

show abstract

Section: Definic ¸ãO De Grafos E Invariantesunclassified

unclassified

Uma análise da relação entre invariantes de grafos e coeficiente de bloqueio em Redes Ópticas Elásticas

Wenzel¹,

Neris²,

Paiva³

et al. 2021

Anais Do XXXIX Simpósio Brasileiro De Redes De Computadores E Sistemas Distribuídos (SBRC 2021)

View full text Add to dashboard Cite

show abstract

“…R 2 is a coefficient of similarity between the calculated and observed values, which tends to unity when calculated value approaches the observed value. The 'n' in equations (5) and 6is just an index and N is the total number of pixels of training data-set.…”

Section: B Cascade Feed Forward Neural Network (Cffnn)mentioning

confidence: 99%

“…The major issue occurs during the analysis is clustering in the concentrated area [4]. Clustering is a famous autonomous motif analysis method which divides the group of n substances into k (>1) sets depending upon coincidences and non-coincidence metric where the value of k can either be pre-determined or not [5]. This paper explains the analysis and categorization utensil of the SAR image.…”

Section: Introductionmentioning

confidence: 99%

Pixel Based Sar Image Classification using Random Forest Algorithm

Balnarsaiah¹,

Prasad²,

Laxminarayana³

2019

IJITEE

View full text Add to dashboard Cite

Synthetic Aperture Radar (SAR) images (Microwave data) were classified using Multi-Layer Feed Forward, Cascade Forward Neural Networks and Random Forest (RF) algorithms. For the Random Forest, a general model for classification of Remotely Sensed Radar dual-polarization data based on RF is implemented and classified of SAR image (microwave data) classifications. The RF model exploits spatial context between neighbouring pixels in an image, and temporal class dependencies between different images of the same region, in the case of multi-temporal data. Based on the well-founded experimental on basis of random forest techniques for classification tasks and the encouraging experimental results in RF algorithm , the authors conclude that the proposed RF algorithm is useful for classification of SAR (Sentinel 1A) imagery and evaluate its accuracy and kappa coefficient.

show abstract

“…The advent of the information theory is indebted to the work carried out by Shannon in the 5th decade of the 20th century. By employing the concepts related to the probability theory, the information theory is used to build a quantitative index for measuring the information content obtained through the observation of a signal [20]; in this way, the capacities of this theory in various domains are illustrated [21][22][23][24][25]. With the advent of the information theory, quantitative modeling of human capacity in information processing was facilitated, and the corresponding results can be explained in the form of the Hick-Hyman law and Fitts' law.…”

Section: Literature Reviewmentioning

confidence: 99%

Pilot workload assessment under different levels of autopilot failure

2018

View full text Add to dashboard Cite

One of the most interesting topics in the eld of human-machine interaction is workload. In this paper, by using information theory concepts, the baud rates generated in all subsystems of a generic simulator of piloting tasks were calculated; then, a unique numerical index, which determines an estimate of the overall workload, was extracted. To examine the e ectiveness of the presented criteria, three tests with di erent levels of autopilot failure were designed in which existing workload was labeled based on the assumed baud rates. A group of subjects carried out these tests as pilots while recording their own ideas about perceived workload. Results con rmed that there were statistically signi cant di erences between the averages of scores assigned by subjects to the overall workload at three levels of di culty. Consequently, the proposed quantitative index was found e ective enough to determine workload levels in the simulator environment, thus facilitating the creation of the needed scenario noticeably.

show abstract

Feature selection for optical network design via a new mutual information estimator

Cited by 9 publications

References 26 publications

Uma análise da relação entre invariantes de grafos e coeficiente de bloqueio em Redes Ópticas Elásticas

Uma análise da relação entre invariantes de grafos e coeficiente de bloqueio em Redes Ópticas Elásticas

Pixel Based Sar Image Classification using Random Forest Algorithm

Pilot workload assessment under different levels of autopilot failure

Contact Info

Product

Resources

About