Hybrid Neural Network Model for Protection of Dynamic Cyber Infrastructure

2019 ◽  
Vol 22 (4) ◽  
pp. 375-382
Author(s):  
Maxim Kalinin ◽  
Roman Demidov ◽  
Peter Zegzhda
Keyword(s):  
Binary Code ◽  
Mesh Networks ◽  
Dynamic Network ◽  
Neural Model ◽  
Data Sets ◽  
Intrusion Prevention ◽  
Software Vulnerabilities ◽  
Architectural Features ◽  
Universal Approach ◽  
Node Isolation

The paper considers a combination of modern artificial neural networks (ANN) that solves the security relative task of intrusion prevention and vulnerabilities detection in cybernetic infrastructure with dynamic network topology. Self-organizing networks, WSN, m2m networks, IIoT, mesh networks are faced with the cyberthreats of specific character: dynamic routing failures, node isolation, DDoS attacks, traffic lack, etc. Most of them are caused by cybersecurity weaknesses: the software vulnerabilities and architectural features of dynamically reconfigured network. The existing methods of binary code analysis and intrusion detection can work with a small number of data sets, are designed for either code inspection or network checking, and are targeted for static networks with regular topology. The proposed neural model demonstrates an universal approach that deals with the cybersecurity weakness as a systems genuine property and attempts to approximate it using a hybrid deep ANN. The new ANN detects both the network security defects and binary code vulnerabilities at once with high accuracy (more than 0.97). It also shows good performance capacity processing big data of the undercontrolled network.

Analysis and Integration of Biological Data

2010 ◽  
pp. 287-314
Author(s):  
Diego Milone ◽  
Georgina Stegmayer ◽  
Matías Gerard ◽  
Laura Kamenetzky ◽  
Mariana López ◽  
...  
Keyword(s):  
Data Integration ◽  
New Technologies ◽  
Biological Significance ◽  
Neural Model ◽  
Biological Data ◽  
Data Sets ◽  
Clustering Methods ◽  
Gene Expressions ◽  
Internal Coherence ◽  
Biological Data Integration

The volume of information derived from post genomic technologies is rapidly increasing. Due to the amount of involved data, novel computational methods are needed for the analysis and knowledge discovery into the massive data sets produced by these new technologies. Furthermore, data integration is also gaining attention for merging signals from different sources in order to discover unknown relations. This chapter presents a pipeline for biological data integration and discovery of a priori unknown relationships between gene expressions and metabolite accumulations. In this pipeline, two standard clustering methods are compared against a novel neural network approach. The neural model provides a simple visualization interface for identification of coordinated patterns variations, independently of the number of produced clusters. Several quality measurements have been defined for the evaluation of the clustering results obtained on a case study involving transcriptomic and metabolomic profiles from tomato fruits. Moreover, a method is proposed for the evaluation of the biological significance of the clusters found. The neural model has shown a high performance in most of the quality measures, with internal coherence in all the identified clusters and better visualization capabilities.

scholarly journals What You Say and How You Say it: Joint Modeling of Topics and Discourse in Microblog Conversations

2019 ◽  
Vol 7 ◽  
pp. 267-281 ◽  
Author(s):  
Jichuan Zeng ◽  
Jing Li ◽  
Yulan He ◽  
Cuiyun Gao ◽  
Michael R. Lyu ◽  
...  
Keyword(s):  
Neural Model ◽  
Joint Modeling ◽  
Data Sets

This paper presents an unsupervised framework for jointly modeling topic content and discourse behavior in microblog conversations. Concretely, we propose a neural model to discover word clusters indicating what a conversation concerns (i.e., topics) and those reflecting how participants voice their opinions (i.e., discourse).1Extensive experiments show that our model can yield both coherent topics and meaningful discourse behavior. Further study shows that our topic and discourse representations can benefit the classification of microblog messages, especially when they are jointly trained with the classifier.Our data sets and code are available at: http://github.com/zengjichuan/Topic_Disc .

BOF Endpoint Prediction Based on Multi-Neural Network Model

2013 ◽  
Vol 441 ◽  
pp. 666-669 ◽  
Author(s):  
You Jun Yue ◽  
Ying Dong Yao ◽  
Hui Zhao ◽  
Hong Jun Wang
Keyword(s):  
Neural Network ◽  
Network Model ◽  
Neural Network Model ◽  
Large Data ◽  
Neural Model ◽  
Large Data Sets ◽  
Data Sets ◽  
Control Precision ◽  
Detection Technology ◽  
Multi Neural Network

In order to solve the problem that the small and middle converters unable to introduce the sublance detection technology to improve the control precision of endpoint because of the constraints of economy and technology, a method which combine the pedigree cluster and neural network is studied, the pedigree cluster divide the large data sets into several categories, the degree of similarity will be relatively high in each category after division, then train neural model for every category. Finally make predictions. Simulation results show that the multi-neural network model has better prediction results.

scholarly journals Phenomenal: An automatic open source library for 3D shoot architecture reconstruction and analysis for image-based plant phenotyping

2019 ◽  
Author(s):  
Simon Artzet ◽  
Tsu-Wei Chen ◽  
Jérôme Chopard ◽  
Nicolas Brichet ◽  
Michael Mielewczik ◽  
...  
Keyword(s):  
Open Source ◽  
High Throughput ◽  
Synthetic Data ◽  
3D Models ◽  
Plant Phenotyping ◽  
Published Data ◽  
Data Sets ◽  
Quantitative Measurements ◽  
Workflow System ◽  
Architectural Features

AbstractIn the era of high-throughput visual plant phenotyping, it is crucial to design fully automated and flexible workflows able to derive quantitative traits from plant images. Over the last years, several software supports the extraction of architectural features of shoot systems. Yet currently no end-to-end systems are able to extract both 3D shoot topology and geometry of plants automatically from images on large datasets and a large range of species. In particular, these software essentially deal with dicotyledons, whose architecture is comparatively easier to analyze than monocotyledons. To tackle these challenges, we designed the Phenomenal software featured with: (i) a completely automatic workflow system including data import, reconstruction of 3D plant architecture for a range of species and quantitative measurements on the reconstructed plants; (ii) an open source library for the development and comparison of new algorithms to perform 3D shoot reconstruction and (iii) an integration framework to couple workflow outputs with existing models towards model-assisted phenotyping. Phenomenal analyzes a large variety of data sets and species from images of high-throughput phenotyping platform experiments to published data obtained in different conditions and provided in a different format. Phenomenal has been validated both on manual measurements and synthetic data simulated by 3D models. It has been also tested on other published datasets to reproduce a published semi-automatic reconstruction workflow in an automatic way. Phenomenal is available as an open-source software on a public repository.

AntMeshNet

2014 ◽  
Vol 5 (1) ◽  
pp. 20-45 ◽  
Author(s):  
Sharad Sharma ◽  
Shakti Kumar ◽  
Brahmjit Singh
Keyword(s):  
Foraging Behavior ◽  
Network Architecture ◽  
Network Performance ◽  
Mesh Networks ◽  
Fault Tolerant ◽  
Dynamic Environment ◽  
Dynamic Network ◽  
Residual Energy ◽  
Network Resource ◽  
Self Organizing

Wireless Mesh Networks (WMNs) are emerging as evolutionary self organizing networks to provide connectivity to end users. Efficient Routing in WMNs is a highly challenging problem due to existence of stochastically changing network environments. Routing strategies must be dynamically adaptive and evolve in a decentralized, self organizing and fault tolerant way to meet the needs of this changing environment inherent in WMNs. Conventional routing paradigms establishing exact shortest path between a source-terminal node pair perform poorly under the constraints imposed by dynamic network conditions. In this paper, the authors propose an optimal routing approach inspired by the foraging behavior of ants to maximize the network performance while optimizing the network resource utilization. The proposed AntMeshNet algorithm is based upon Ant Colony Optimization (ACO) algorithm; exploiting the foraging behavior of simple biological ants. The paper proposes an Integrated Link Cost (ILC) measure used as link distance between two adjacent nodes. ILC takes into account throughput, delay, jitter of the link and residual energy of the node. Since the relationship between input and output parameters is highly non-linear, fuzzy logic was used to evaluate ILC based upon four inputs. This fuzzy system consists of 81 rules. Routing tables are continuously updated after a predefined interval or after a change in network architecture is detected. This takes care of dynamic environment of WMNs. A large number of trials were conducted for each model. The results have been compared with Adhoc On-demand Distance Vector (AODV) algorithm. The results are found to be far superior to those obtained by AODV algorithm for the same WMN.

Routing in Wireless Mesh Networks based on Termites' Intelligence

2017 ◽  
Vol 8 (2) ◽  
pp. 1-21 ◽  
Author(s):  
Sharad Sharma ◽  
Asha Malik
Keyword(s):  
High Speed ◽  
Mesh Networks ◽  
Optimal Path ◽  
Dynamic Network ◽  
Critical Issue ◽  
Mesh Network ◽  
Wireless Mesh ◽  
Heuristic Technique ◽  
Link Cost ◽  
Selection Of

Wireless Mesh Network (WMN) is envisaged to be key component of next generation wireless networks which can effectively cope with the ever increasing fast and vast growing need to access data and avail services over the network. These networks provide infrastructure less high speed internet access to the end users and have a cutting edge over the existing networks. Routing, being most critical issue in their implementation. The dynamic network conditions impose setbacks in the selection of optimal path. There is exigent requirement to tackle these routing issues in context of these networks. In this paper, the authors apply a nature inspired soft computing based meta heuristic technique called Termite Colony Optimization in WMN to find an optimal route based on the link cost. TCO approach is inspired by the emergent behaviour exhibited by the natural termite colony swarms for mound building. Experimentally TCO shows faster converges over some existing algorithms.

Community Fabric: Visualizing communities and structure in dynamic networks

2021 ◽  
pp. 147387162110560
Author(s):  
Evan Ezell ◽  
Seung-Hwan Lim ◽  
David Anderson ◽  
Robert Stewart
Keyword(s):  
Dynamic Networks ◽  
Dynamic Network ◽  
Data Sets ◽  
Dynamic Graphs ◽  
Visualization Technique ◽  
Web Based ◽  
New Approach ◽  
Community Evolution ◽  
Underlying Network ◽  
Flow Diagrams

We present Community Fabric, a novel visualization technique for simultaneously visualizing communities and structure within dynamic networks. In dynamic networks, the structure of the network is continuously evolving throughout time and these underlying topological shifts tend to lead to communal changes. Community Fabric helps the viewer more easily interpret and understand the interplay of structural change and community evolution in dynamic graphs. To achieve this, we take a new approach, hybridizing two popular network and community visualizations. Community Fabric combines the likes of the Biofabric static network visualization method with traditional community alluvial flow diagrams to visualize communities in a dynamic network while also displaying the underlying network structure. Our approach improves upon existing state-of-the-art techniques in several key areas. We describe the methodologies of Community Fabric, implement the visualization using modern web-based tools, and apply our approach to three example data sets.

Analysis and Integration of Biological Data

Data Mining ◽  
2013 ◽  
pp. 203-230
Author(s):  
Diego Milone ◽  
Georgina Stegmayer ◽  
Matías Gerard ◽  
Laura Kamenetzky ◽  
Mariana López ◽  
...  
Keyword(s):  
Data Integration ◽  
New Technologies ◽  
Biological Significance ◽  
Neural Model ◽  
Biological Data ◽  
Data Sets ◽  
Clustering Methods ◽  
Gene Expressions ◽  
Internal Coherence ◽  
Biological Data Integration

The volume of information derived from post genomic technologies is rapidly increasing. Due to the amount of involved data, novel computational methods are needed for the analysis and knowledge discovery into the massive data sets produced by these new technologies. Furthermore, data integration is also gaining attention for merging signals from different sources in order to discover unknown relations. This chapter presents a pipeline for biological data integration and discovery of a priori unknown relationships between gene expressions and metabolite accumulations. In this pipeline, two standard clustering methods are compared against a novel neural network approach. The neural model provides a simple visualization interface for identification of coordinated patterns variations, independently of the number of produced clusters. Several quality measurements have been defined for the evaluation of the clustering results obtained on a case study involving transcriptomic and metabolomic profiles from tomato fruits. Moreover, a method is proposed for the evaluation of the biological significance of the clusters found. The neural model has shown a high performance in most of the quality measures, with internal coherence in all the identified clusters and better visualization capabilities.

Convergecast Tree on Temporal Graphs

2020 ◽  
Vol 31 (03) ◽  
pp. 385-409
Author(s):  
Subhrangsu Mandal ◽  
Arobinda Gupta
Keyword(s):  
Data Collection ◽  
Heuristic Algorithm ◽  
Arrival Time ◽  
Real Life ◽  
Dynamic Network ◽  
Data Sets ◽  
Real Life Data ◽  
Network Topologies ◽  
Dynamic Sensor Networks ◽  
Temporal Graphs

Temporal graphs are useful tools to model dynamic network topologies found in many applications. In this paper, we address the problem of constructing a convergecast tree on temporal graphs for data collection in dynamic sensor networks. Two types of convergecast trees, bounded arrival time convergecast tree and minimum total arrival time convergecast tree are defined as useful structures for low delay data collection. An [Formula: see text] time centralized algorithm is proposed to construct a bounded arrival time convergecast tree, where [Formula: see text] is the number of nodes, [Formula: see text] is the number of edges, and [Formula: see text] is the lifetime of the given temporal graph. The algorithm presented is an offline algorithm and assumes that all information about change in the graph topology is known apriori. It is then shown that the problem of constructing a minimum total arrival time convergecast tree is NP-complete, and an [Formula: see text] time centralized, offline heuristic algorithm is proposed to address it. The heuristic algorithm is evaluated with experiments on four real life data sets.

Sign in / Sign up

Export Citation Format

Share Document