scholarly journals Information Bottleneck Classification in Extremely Distributed Systems

Entropy ◽  
2020 ◽  
Vol 22 (11) ◽  
pp. 1237
Author(s):  
Denis Ullmann ◽  
Shideh Rezaeifar ◽  
Olga Taran ◽  
Taras Holotyak ◽  
Brandon Panos ◽  
...  
Keyword(s):  
Big Data ◽  
Probability Distribution ◽  
Classification System ◽  
Data Communication ◽  
Training Dataset ◽  
Final Decision ◽  
Communication Problems ◽  
Information Bottleneck ◽  
Class Probability ◽  
Distributed Scheme

We present a new decentralized classification system based on a distributed architecture. This system consists of distributed nodes, each possessing their own datasets and computing modules, along with a centralized server, which provides probes to classification and aggregates the responses of nodes for a final decision. Each node, with access to its own training dataset of a given class, is trained based on an auto-encoder system consisting of a fixed data-independent encoder, a pre-trained quantizer and a class-dependent decoder. Hence, these auto-encoders are highly dependent on the class probability distribution for which the reconstruction distortion is minimized. Alternatively, when an encoding–quantizing–decoding node observes data from different distributions, unseen at training, there is a mismatch, and such a decoding is not optimal, leading to a significant increase of the reconstruction distortion. The final classification is performed at the centralized classifier that votes for the class with the minimum reconstruction distortion. In addition to the system applicability for applications facing big-data communication problems and or requiring private classification, the above distributed scheme creates a theoretical bridge to the information bottleneck principle. The proposed system demonstrates a very promising performance on basic datasets such as MNIST and FasionMNIST.

scholarly journals Intrusion Detection Based on Big Data Fuzzy Analytics

2021 ◽  
Author(s):  
Farah Jemili ◽  
Hajer Bouras
Keyword(s):  
Big Data ◽  
Network Security ◽  
Intrusion Detection ◽  
Intrusion Detection System ◽  
High Performance ◽  
Detection System ◽  
Training Dataset ◽  
False Alarms ◽  
Massive Datasets ◽  
Detection Rates

In today’s world, Intrusion Detection System (IDS) is one of the significant tools used to the improvement of network security, by detecting attacks or abnormal data accesses. Most of existing IDS have many disadvantages such as high false alarm rates and low detection rates. For the IDS, dealing with distributed and massive data constitutes a challenge. Besides, dealing with imprecise data is another challenge. This paper proposes an Intrusion Detection System based on big data fuzzy analytics; Fuzzy C-Means (FCM) method is used to cluster and classify the pre-processed training dataset. The CTU-13 and the UNSW-NB15 are used as distributed and massive datasets to prove the feasibility of the method. The proposed system shows high performance in terms of accuracy, precision, detection rates, and false alarms.

scholarly journals Brazilian Review of Finance 2015 Editorial Report

2016 ◽  
Vol 14 (1) ◽  
pp. 1
Author(s):  
Marcio Poletti Laurini
Keyword(s):  
Classification System ◽  
Acceptance Rate ◽  
Google Scholar ◽  
Final Decision ◽  
Editorial Process ◽  
Worst Case ◽  
Publication Schedule ◽  
Brazilian Publication

RBFin is the main Brazilian publication outlet of academic papers about finance. Using the Open Journals System to manage the editorial process, publication of RBFin adheres to a strict publication schedule. The journal is indexed by EconLit, RedALyC, Proquest, Google Scholar, Gale and Ebsco and is listed in the JEL, DOAJ, Latindex, OpenJGate, and Cabell’s directories. RBFin is rated B2 in the business and economics areas of the Brazilian classification system. The editorial board undergoes partial turnover every year and comprises 19 individuals from four countries, the Brazilian members being affiliated with universities in five different Brazilian states. The acceptance rate was 44\% for papers submitted in 2015. The average number of days between receipt and first decision for articles submitted in 2015 was 86. The average number of days between receipt and final decision for articles submitted in 2015 was 104. The worst case was 345 days. Thirty five individuals served as reviewers in 2015.

Regional seismic risk in British Columbia — classification of buildings and development of damage probability functions

2005 ◽  
Vol 32 (2) ◽  
pp. 372-387 ◽  
Author(s):  
Carlos E Ventura ◽  
W.D Liam Finn ◽  
Tuna Onur ◽  
Ardel Blanquera ◽  
Mahmoud Rezai
Keyword(s):  
British Columbia ◽  
Probability Distribution ◽  
Classification System ◽  
Seismic Risk ◽  
Distribution Functions ◽  
Intensity Level ◽  
Ground Shaking ◽  
Damage Level ◽  
Damage Probability ◽  
Damage Probability Matrix

Regional seismic risk estimations are needed in southwestern British Columbia, since it is one of the most seismically active and highly populated regions in Canada. Regional estimations typically involve a large number of buildings, which makes it necessary to establish a building classification system, where the average response to earthquake shaking is assumed to be similar within each building class. In this study, buildings in British Columbia were divided into 31 classes based on their material, lateral load bearing system, height, use, and age. A damage probability matrix (DPM) was then developed for each building class which describes the probability of being in a certain damage level (i.e., light, moderate, heavy, etc.) given the ground shaking intensity. Next, a probability distribution function was fit to the discrete probability values at each intensity level. The products of this study, the building classification system, the DPMs, and the probability distribution functions will allow regional damage and loss estimations in the area.Key words: seismic risk, vulnerability, building classification, structural system, building response, damage, probability.

Gaussian Process Emulation for Big Data in Data-Driven Metamaterials Design

2019 ◽  
Author(s):  
Ramin Bostanabad ◽  
Yu-Chin Chan ◽  
Liwei Wang ◽  
Ping Zhu ◽  
Wei Chen
Keyword(s):  
Big Data ◽  
Gaussian Process ◽  
Engineering Design ◽  
Training Data ◽  
Data Driven ◽  
Training Dataset ◽  
Massive Datasets ◽  
Unit Cells ◽  
Gaussian Process Emulation ◽  
Novel Method

Abstract Our main contribution is to introduce a novel method for Gaussian process (GP) modeling of massive datasets. The key idea is to build an ensemble of independent GPs that use the same hyperparameters but distribute the entire training dataset among themselves. This is motivated by our observation that estimates of the GP hyperparameters change negligibly as the size of the training data exceeds a certain level, which can be found in a systematic way. For inference, the predictions from all GPs in the ensemble are pooled to efficiently exploit the entire training dataset for prediction. We name our modeling approach globally approximate Gaussian process (GAGP), which, unlike most largescale supervised learners such as neural networks and trees, is easy to fit and can interpret the model behavior. These features make it particularly useful in engineering design with big data. We use analytical examples to demonstrate that GAGP achieves very high predictive power that matches or exceeds that of state-of-the-art machine learning methods. We illustrate the application of GAGP in engineering design with a problem on data-driven metamaterials design where it is used to link reduced-dimension geometrical descriptors of unit cells and their properties. Searching for new unit cell designs with desired properties is then accomplished by employing GAGP in inverse optimization.

A novel prediction model of traffic accidents based on big data

2019 ◽  
Vol 10 (04) ◽  
pp. 1950022 ◽  
Author(s):  
Minglei Song ◽  
Rongrong Li ◽  
Binghua Wu
Keyword(s):  
Feature Extraction ◽  
Big Data ◽  
Probability Distribution ◽  
Prediction Model ◽  
Probability Density ◽  
Traffic Accidents ◽  
Joint Probability ◽  
Distributed Database ◽  
Joint Probability Distribution ◽  
Joint Probability Density

The occurrence of traffic accidents is regular in probability distribution. Using big data mining method to predict traffic accidents is conducive to taking measures to prevent or reduce traffic accidents in advance. In recent years, prediction methods of traffic accidents used by researchers have some problems, such as low calculation accuracy. Therefore, a prediction model of traffic accidents based on joint probability density feature extraction of big data is proposed in this paper. First, a function of big data joint probability distribution for traffic accidents is established. Second, establishing big data distributed database model of traffic accidents with the statistical analysis method in order to mine the association rules characteristic quantity reflecting the law of traffic accidents, and then extracting the joint probability density feature of big data for traffic accident probability distribution. According to the result of feature extraction, adaptive functional and directivity are predicted, and then the regularity prediction of traffic accidents is realized based on the result of association directional clustering, so as to optimize the design of the prediction model of traffic accidents based on big data. Simulation results show that in predicting traffic accidents, the model in this paper has advantages of relatively high accuracy, relatively good confidence and stable prediction result.

scholarly journals Relatório Editorial de 2012 da Revista Brasileira de Finanças

2013 ◽  
Vol 11 (1) ◽  
pp. 1
Author(s):  
Ricardo Pereira Câmara Leal
Keyword(s):  
Classification System ◽  
Acceptance Rate ◽  
Google Scholar ◽  
Final Decision ◽  
Editorial Process ◽  
Worst Case ◽  
Business Area ◽  
Publication Schedule ◽  
Brazilian Publication

RBFin is the main Brazilian publication outlet of academic papers about finance. The contents of the Review are open and online; a printed version was maintained, in part, thanks to a grant from CNPq/CAPES. Using the Open Journals System to manage the editorial process, publication of RBFin adheres to a strict publication schedule. The journal is indexed by EconLit, Google Scholar, Gale and Ebsco and is listed in the JEL, DOAJ, Latindex, OpenJGate, and Cabell's directories. It will soon appear in RedALyC and Proquest. RBFin is rated B1 in the business area of the Brazilian classification system. The editorial board undergoes partial turnover every year and comprises 18 individuals from four countries, the Brazilian members being affiliated with universities in five different Brazilian states. The acceptance rate was 18% for papers submitted in 2011, the most recent year in which all submissions have already received a final decision. The average number of days between receipt and acceptance for articles submitted in 2012 was 215. The worst case was 584 days. The average number of days between receipt and publication was 317. The worst case was 760 days. Sixty-three individuals served as reviewers in 2012.

Sign in / Sign up

Export Citation Format

Share Document