 |
Neural Networks 2015 |
Artificial neural networks have been used to solve a wide variety of tasks that are hard to solve using ordinary rule-based programming. What has attracted much interest in neural networks is the possibility of learning. Tasks such as function approximation, classification pattern and sequence recognition, anomaly detection, filtering, clustering, blind source separation and compression and controls all have security implications. For the Science of Security community, neural network research is related to metrics, resilience, and privacy. The work cited here was presented in 2015.
Sagar, V.; Kumar, K., "A symmetric key cryptography using genetic algorithm and error back propagation neural network," in Computing for Sustainable Global Development (INDIACom), 2015 2nd International Conference on, pp. 1386-1391, 11-13 March 2015
Abstract: In conventional security mechanism, cryptography is a process of information and data hiding from unauthorized access. It offers the unique possibility of certifiably secure data transmission among users at different remote locations. Cryptography is used to achieve availability, privacy and integrity over different networks. Usually, there are two categories of cryptography i.e. symmetric and asymmetric. In this paper, we have proposed a new symmetric key algorithm based on genetic algorithm (GA) and error back propagation neural network (EBP-NN). Genetic algorithm has been used for encryption and neural network has been used for decryption process. Consequently, this paper proposes an easy cryptographic secure algorithm for communication over the public computer networks.
keywords: backpropagation;computer network security;cryptography;genetic algorithms;neural nets;EBP-NN;GA;certifiably secure data transmission;cryptographic secure algorithm;data hiding;data integrity;data privacy;decryption process;error back propagation neural network;genetic algorithm;information hiding;public computer networks;remote locations;symmetric key cryptography;unauthorized access;Artificial neural networks;Encryption;Genetic algorithms;Neurons;Receivers;cryptography;error back propagation neural network;genetic algorithm;symmetric key (ID#: 16-9253)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7100476&isnumber=7100186
Sujatha, K.; Nageswara Rao, P.V.; Rao, A.A.; Prasad, K.R.; Deepthi, M.S.B., "Biometric Identity Verification using Automatic Speaker Recognition," in Electrical, Electronics, Signals, Communication and Optimization (EESCO), 2015 International Conference on, pp. 1-5, 24-25 Jan. 2015. doi: 10.1109/EESCO.2015.7253813
Abstract: Password based Security Systems have to provide Authentication and Privacy. Strong privacy protected and high security authentication System design remains as an open problem because of the weak passwords. The fundamental problem in normal text-based passwords is that weak passwords are vulnerable to attacks and strong passwords creating problem for the user as he has to remember them with difficulty. Biometric features like passwords uttered by speakers are well suited for authentication, as this cannot be stolen or recreated and is unique to each individual. Biometric Identity Verification using Automatic Speaker Recognition is the neural network based process of automatically recognizing the user from a recording, using the text-dependent password uttered by speech. Here, Biometric identity verification is the speech that can be used to either accept or reject the identity claimed by a given user. Text supports user to remember the password when he forgets and also even when revealed to others it does not harm the system as it takes only voice password from authorized user. Voice is considered a valuable biometric feature which depends on the specific person's speaking style and physical attributes, and also it is very easy to collect and process speech data. The capacity of proposed security framework is to have a framework that will just open after perceiving a voice secret word talked by the watchword holder utilizing an Artificial Neural Network.
keywords: biometrics (access control);neural nets;speaker recognition;artificial neural network;automatic speaker recognition;biometric identity verification;password based security systems;Authentication;Biological system modeling;Mel frequency cepstral coefficient;Speaker recognition;Speech;Speech recognition;Artificial Neural Network (ANN);Automatic Speaker Recognition (ASR);Biometric Identity Verification (BIV) (ID#: 16-9254)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7253813&isnumber=7253613
Ashwin Kumar, T.K.; Hong Liu; Thomas, J.P.; Mylavarapu, G., "Identifying Sensitive Data Items within Hadoop," in High Performance Computing and Communications (HPCC), 2015 IEEE 7th International Symposium on Cyberspace Safety and Security (CSS), 2015 IEEE 12th International Conference on Embedded Software and Systems (ICESS), 2015 IEEE 17th International Conference on, pp. 1308-1313, 24-26 Aug. 2015. doi: 10.1109/HPCC-CSS-ICESS.2015.293
Abstract: Recent growth in big-data is raising security and privacy concerns. Organizations that collect data from various sources are at a risk of legal or business liabilities due to security breach and exposure of sensitive information. Only file-level access control is feasible in current Hadoop implementation and the sensitive information can only be identified manually or from the information provided by the data owner. The problem of identifying sensitive information manually gets complicated due to different types of data. When sensitive information is accessed by an unauthorized user or misused by an authorized person, they can compromise privacy. This paper is the first part of our intended access control framework for Hadoop and it automates the process of identifying sensitive data items manually. To identify such data items, the proposed framework harnesses data context, usage patterns and data provenance. In addition to this the proposed framework can also keep track of the data lineage.
keywords: Big Data;authorisation;data handling;data privacy;parallel processing;Big-Data;Hadoop;access control framework;authorized person;business liabilities;data collection;data context;data lineage;data privacy;data provenance;data security;file-level access control;information misuse;legal liabilities;security breach;sensitive data item identification;sensitive information access;sensitive information exposure;sensitive information identification;unauthorized user;usage patterns;Access control;Context;Electromyography;Generators;Metadata;Neural networks;Sensitivity;Hadoop;data context;data lineage;data provenance;file-level access control;privacy;sensitive information;usage patterns (ID#: 16-9255)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7336348&isnumber=7336120
Randazzo, F.; Croce, D.; Tinnirello, I.; Barcellona, C.; Merani, M.L., "Experimental evaluation of privacy-preserving aggregation schemes on planetlab," in Wireless Communications and Mobile Computing Conference (IWCMC), 2015 International, pp. 379-384, 24-28 Aug. 2015. doi: 10.1109/IWCMC.2015.7289113
Abstract: New pervasive technologies often reveal many sensitive information about users' habits, seriously compromising the privacy and sometimes even the personal security of people. To cope with this problem, researchers have developed the idea of privacy-preserving data mining which refers to the possibility of releasing aggregate information about the data provided by multiple users, without any information leakage about individual data. These techniques have different privacy levels and communication costs, but all of them can suffer when some users' data becomes inaccessible during the operation of the privacy preserving protocols. It is thus interesting to validate the applicability of such architectures in real-world scenarios. In this paper we experimentally evaluate two promising privac-preserving techniques on PlanetLab, analyzing the execution time and the failure rate that each scheme exhibits.
keywords: data mining;data privacy;ubiquitous computing;PlanetLab;communication costs;pervasive technologies;privacy preserving protocols;privacy-preserving aggregation schemes;privacy-preserving data mining;Artificial neural networks;Cryptography;Data privacy;Peer-to-peer computing;Protocols;Servers;data mining;privacy;secret sharing;secure multi-party computation (ID#: 16-9256)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7289113&isnumber=7288920
Beiye Liu; Chunpeng Wu; Hai Li; Yiran Chen; Qing Wu; Barnell, M.; Qinru Qiu, "Cloning your mind: Security challenges in cognitive system designs and their solutions," in Design Automation Conference (DAC), 2015 52nd ACM/EDAC/IEEE, pp. 1-5, 8-12 June 2015. doi: 10.1145/2744769.2747915
Abstract: With the booming of big-data applications, cognitive information processing systems that leverage advanced data processing technologies, e.g., machine learning and data mining, are widely used in many industry fields. Although these technologies demonstrate great processing capability and accuracy in the relevant applications, several security and safety challenges are also emerging against these learning based technologies. In this paper, we will first introduce several security concerns in cognitive system designs. Some real examples are then used to demonstrate how the attackers can potentially access the confidential user data, replicate a sensitive data processing model without being granted the access to the details of the model, and obtain some key features of the training data by using the services publically accessible to a normal user. Based on the analysis of these security challenges, we also discuss several possible solutions that can protect the information privacy and security of cognitive systems during different stages of the usage.
keywords: Big Data;cognition;security of data;Big-Data application;cognitive information processing systems;cognitive system design;data mining;data security;machine learning;sensitive data processing model;Data models;Neural networks;Predictive models;Security;Training;Training data;Cognitive Systems;Machine Learning;Security (ID#: 16-9257)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7167279&isnumber=7167177
[Title page]," in Futuristic Trends on Computational Analysis and Knowledge Management (ABLAZE), 2015 International Conference on, pp. 1-1, 25-27 Feb. 2015. doi: 10.1109/ABLAZE.2015.7155050
Abstract: The following topics are dealt with: vibration signal based monitoring; mechanical microdrilling; rule based inflectional urdu stemmer usal; rule based derivational urdu stemmer usal; fuzzy logic controller; heat exchanger temperature process; text dependent speaker recognition; MFCC; SBC; multikeyword based sorted querying; encrypted cloud data; communication understandability enhancement; GSD; parsing; input power quality; switched reluctance motor drive; externally powered upper limb prostheses; program test data generation; launch vehicle optimal trajectory generation; misalignment fault detection; induction motors; current signature analysis; vibration signature analysis; wind power plants; vortex induced vibration; mechanical structure modal analysis; machining parameter optimization; diesel engines; high speed nonvolatile NEMS memory devices; image fusion; RGB color space; LUV color space; offline English character recognition; human skin detection; tumor boundary extraction; MR images; OdiaBraille; text transcription; shadow detection; YIQ color models; color aerial images; moving object segmentation; image data deduplication; iris recognition; two-stage series connected thermoelectric generator; education information system; cyclone separator CFD simulation; imperfect debugging; vulnerability discovery model; stochastic differential equation; cloud data access; attribute based encryption; agile SCRUM framework; PID controller optimisation; hybrid watermarking technique; privacy preservation; vertical partitioned medical database; power amplifier; software reliability growth modeling; cochlear implantation; cellular towers; feedforward neural networks; MBSOM; agent based semantic ontology matching; phonetic word identification; test case selection; MANET security issues; online movie data classification; modified LEACH protocol; mobile ad hoc networks; virtual machine introspection; task scheduling; cluster computing; image compression; green cloud computin- ; critical health data transmission system; irreversible regenerative Brayton cycle; task set based adaptive round robin scheduling; database security; heterogeneous online social networks; aspect oriented systems; IP network; MPLS network; DBSCAN algorithm; VANET; self-organizing feature map; image segmentation; enzyme classification; wireless sensor networks; energy smart routing protocol; adaptive gateway discovery mechanism; heuristic job scheduling; AODV based congestion control protocol; expert system; home appliances; relay node based heer protocol; data storage; TORA security; data aggregation; low energy adaptive stable energy efficient protocol; fuzzy logic based clustering algorithm; hybrid evolutionary MPLS tunneling algorithm; English mobile teaching; eigenvector centrality; genetic algorithms; data mining; heart disease prediction; lossless data compression; reconfigurable ring resonator; triple band stacked patch antenna; energy based spectrum sensing; cognitive radio networks; FPGA; knowledge representation; multiband microstrip antenna; Web indexing; HTML priority system; Web cache recommender system; e-learning; IT skill learning for visual impaired; user review data analysis; software up-gradation model; software testing; Web crawlers; secret key watermarking; WAV audio file; SRM drive; ZETA converter; fractional PID tuning; medical image reconstruction; speech recognition system; video authentication; digital forensics; content based image retrieval; image classification; hybrid wavelet transform; facial feature extraction; RBSD adder; smart home environment; generalized discrete time model; We Chat marketing; foreign language learning; carbon dioxide emission mitigation; power generation; smartphone storage enhancement; and virtualization.
keywords: Brayton cycle;IP networks;adders;aspect-oriented programming;audio watermarking;biomedical MRI;cardiology;character recognition;cloud computing;cochlear implants;cognitive radio;computational fluid dynamics;computer science education;content-based retrieval;cryptography;cyclone separators;data analysis;data compression;data mining;data privacy;diesel engines;differential equations;digital forensics;domestic appliances;drilling;educational administrative data processing;eigenvalues and eigenfunctions;enzymes;expert systems;face recognition;fault diagnosis;feature extraction;feedforward neural nets;field programmable gate arrays;fuzzy control;genetic algorithms;grammars;green computing;handicapped aids;heat exchangers;home computing;image classification;image coding;image colour analysis;image fusion;image reconstruction;image retrieval;image segmentation;image watermarking;indexing;induction motors;internetworking;iris recognition;knowledge representation;linguistics;medical image processing;microstrip antennas;mobile learning;modal analysis;nanoelectromechanical devices;object detection;ontologies (artificial intelligence);pattern clustering;power amplifiers;power supply quality;program debugging;program testing;radio spectrum management;recommender systems;reluctance motor drives;resonators;routing protocols;scheduling;self-organising feature maps;social networking (online);software reliability;speaker recognition;speech processing;storage management;telecommunication congestion control;thermoelectric conversion;three-term control;trajectory optimisation (aerospace);tumours;vehicular ad hoc networks;vibrations;video signal processing;virtual machines;virtualisation;wavelet transforms;wind power plants;wireless sensor networks;AODV based congestion control protocol;DBSCAN algorithm;English mobile teaching;FPGA;GSD;HTML priority system;IP network;IT skill learning for visual impaired;LUV color space;MANET security issues;MBSOM;MFCC;MPLS network;MR images;OdiaBraille;PID controller optimisation;RBSD adder;RGB color space;SBC;SRM drive;TORA security;VANET;WAV audio file;We Chat marketing;Web cache recommender system;Web crawlers;Web indexing;YIQ color models;ZETA converter;adaptive gateway discovery mechanism;agent based semantic ontology matching;agile SCRUM framework;aspect oriented systems;attribute based encryption;carbon dioxide emission mitigation;cellular towers;cloud data access;cluster computing;cochlear implantation;cognitive radio networks;color aerial images;communication understandability enhancement;content based image retrieval;critical health data transmission system;current signature analysis;cyclone separator CFD simulation;data aggregation;data mining;data storage;database security;diesel engines;digital forensics;e-learning;education information system;eigenvector centrality;encrypted cloud data;energy based spectrum sensing;energy smart routing protocol;enzyme classification;expert system;externally powered upper limb prostheses;facial feature extraction;feedforward neural networks;foreign language learning;fractional PID tuning;fuzzy logic based clustering algorithm;fuzzy logic controller;generalized discrete time model;genetic algorithms;green cloud computing;heart disease prediction;heat exchanger temperature process;heterogeneous online social networks;heuristic job scheduling;high speed nonvolatile NEMS memory devices;home appliances;human skin detection;hybrid evolutionary MPLS tunneling algorithm;hybrid watermarking technique;hybrid wavelet transform;image classification;image compression;image data deduplication;image fusion;image segmentation;imperfect debugging;induction motors;input power quality;iris recognition;irreversible regenerative Brayton cycle;knowledge representation;launch vehicle optimal trajectory generation;lossless data compression;low energy adaptive stable energy efficient protocol;machining parameter optimization;mechanical microdrilling;mechanical structure modal analysis;medical image reconstruction;misalignment fault detection;mobile ad hoc networks;modified LEACH protocol;moving object segmentation;multiband microstrip antenna;multikeyword based sorted querying;offline English character recognition;online movie data classification;parsing;phonetic word identification;power amplifier;power generation;privacy preservation;program test data generation;reconfigurable ring resonator;relay node based heer protocol;rule based derivational urdu stemmer usal;rule based inflectional urdu stemmer usal;secret key watermarking;self-organizing feature map;shadow detection;smart home environment;smartphone storage enhancement;software reliability growth modeling;software testing;software up-gradation model;speech recognition system;stochastic differential equation;switched reluctance motor drive;task scheduling;task set based adaptive round robin scheduling;test case selection;text dependent speaker recognition;text transcription;triple band stacked patch antenna;tumor boundary extraction;two-stage series connected thermoelectric generator;user review data analysis;vertical partitioned medical database;vibration signal based monitoring;vibration signature analysis;video authentication;virtual machine introspection;virtualization;vortex induced vibration;vulnerability discovery model;wind power plants;wireless sensor networks (ID#: 16-9258)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7155050&isnumber=7154914
Vukovic, M.; Skocir, P.; Katusic, D.; Jevtic, D.; Trutin, D.; Delonga, L., "Estimating real world privacy risk scenarios," in Telecommunications (ConTEL), 2015 13th International Conference on, pp. 1-7, 13-15 July 2015. doi: 10.1109/ConTEL.2015.7231214
Abstract: User privacy is becoming an issue on the Internet due to common data breaches and various security threats. Services tend to require private user data in order to provide more personalized content and users are typically unaware of potential risks to their privacy. This paper continues our work on the proposed user privacy risk calculator based on a feedforward neural network. Along with risk estimation, we provide the users with real world example scenarios that depict privacy threats according to selected input parameters. In this paper, we present a model for selecting the most probable real world scenario, presented as a comic, and thus avoid overwhelming the user with lots of information that he/she may find confusing. Most probable scenario estimations are performed by artificial neural network that is trained with real world scenarios and estimated probabilities from real world occurrences. Additionally, we group real world scenarios into categories that are presented to the user as further reading regarding privacy risks.
keywords: data privacy;feedforward neural nets;learning (artificial intelligence);probability;artificial neural network training;data breach;feed-forward neural network;input parameter selection;personalized content;privacy risks;privacy threats;private user data;probabilities;real-world privacy risk scenario estimation;risk estimation;security threats;user privacy;user privacy risk calculator;Calculators;Electronic mail;Estimation;Internet;Law;Privacy (ID#: 16-9259)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7231214&isnumber=7231179
Cuiling Jiang; Yilin Pang; Anwen Wu, "A Novel Robust Image-Hashing Method for Content Authentication," in Security and Privacy in Social Networks and Big Data (SocialSec), 2015 International Symposium on, pp. 22-27, 16-18 Nov. 2015. doi: 10.1109/SocialSec2015.15
Abstract: Image hash functions find extensive application in content authentication, database search, and digital forensic. This paper develops a novel robust image-hashing method based on genetic algorithm (GA) and Back Propagation (BP) Neural Network for content authentication. Lifting wavelet transform is used to extract image low frequency coefficients to create the image feature matrix. A GA-BP network model is constructed to generate image-hashing code. Experimental results demonstrate that the proposed hashing method is robust against random attack, JPEG compression, additive Gaussian noise, and so on. Receiver operating characteristics (ROC) analysis over a large image database reveals that the proposed method significantly outperforms other approaches for robust image hashing.
keywords: Gaussian noise;authorisation;backpropagation;cryptography;data compression;genetic algorithms;image coding;neural nets;sensitivity analysis;wavelet transforms;GA-BP network model;JPEG compression;ROC;additive Gaussian noise;back propagation neural network;content authentication;database search;digital forensic;genetic algorithm;image database;image feature matrix;image hash functions;image low frequency coefficients extract;image-hashing code;lifting wavelet transform;receiver operating characteristics analysis;robust image-hashing method;Authentication;Feature extraction;Genetic algorithms;Robustness;Training;Wavelet transforms;BP network;discrimination;genetic algorithm;image hash (ID#: 16-9260)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7371895&isnumber=7371823
Damasceno, M.; Canuto, A.M.P.; Poh, N., "Multi-privacy biometric protection scheme using ensemble systems," in Neural Networks (IJCNN), 2015 International Joint Conference on, pp. 1-8, 12-17 July 2015. doi: 10.1109/IJCNN.2015.7280657
Abstract: Biometric systems use personal biological or behavioural traits that can uniquely characterise an individual but this uniqueness property also becomes its potential weakness when the template characterising a biometric trait is stolen or compromised. To this end, we consider two strategies to improving biometric template protection and performance, namely, (1) using multiple privacy schemes and (2) using multiple matching algorithms. While multiple privacy schemes can improve the security of a biometric system by protecting its template; using multiple matching algorithms or similarly, multiple biometric traits along with their respective matching algorithms, can improve the system performance due to reduced intra-class variability. The above two strategies lead to a novel, ensemble system that is derived from multiple privacy schemes. Our findings suggest that, under the worst-case scenario evaluation where the key or keys protecting the template are stolen, multi-privacy protection scheme can outperform a single protection scheme as well as the baseline biometric system without template protection.
keywords: biometrics (access control);data privacy;behavioural traits;biological traits;biometric template protection;ensemble systems;multiple matching algorithms;multiprivacy biometric protection;worst-case scenario evaluation;Biology;Training (ID#: 16-9261)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7280657&isnumber=7280295
Boyang Li; Chen Liu, "Parallel BP Neural Network on Single-chip Cloud Computer," in High Performance Computing and Communications (HPCC), 2015 IEEE 7th International Symposium on Cyberspace Safety and Security (CSS), 2015 IEEE 12th International Conference on Embedded Software and Systems (ICESS), 2015 IEEE 17th International Conference on, pp. 1871-1875, 24-26 Aug. 2015. doi: 10.1109/HPCC-CSS-ICESS.2015.280
Abstract: Neural network has been a clear focus in machine learning area. Back propagation (BP) method is frequently used in neural network training. In this work we paralleled BP neural network on Single-Chip Cloud Computer (SCC), an experimental processor created by Intel Labs, and analyzed multiple metrics under different configurations. We also varied the number of neurons (nodes) in the hidden layer of the BP neural networks and studied the impact. The experiment results show that a better performance can be obtained with SCC, especially when there are more nodes in the hidden layer of BP neural network. A low voltage and frequency configuration contributes to a low power per speedup. What is more, a medium voltage and frequency configuration contributes to both a low energy consumption and energy-delay product.
keywords: backpropagation;cloud computing;learning (artificial intelligence);Intel Labs;SCC;back propagation method;machine learning;parallel BP neural network;single-chip cloud computer;Biological neural networks;Computers;Energy consumption;Frequency-domain analysis;Power demand;Training;Back Propagation;Energy-Aware Computing;Neural Netork;Power-Aware Computing;Single-chip Cloud Computer (ID#: 16-9262)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7336445&isnumber=7336120
Manem, H.; Beckmann, K.; Min Xu; Carroll, R.; Geer, R.; Cady, N.C., "An extendable multi-purpose 3D neuromorphic fabric using nanoscale memristors," in Computational Intelligence for Security and Defense Applications (CISDA), 2015 IEEE Symposium on, pp. 1-8, 26-28 May 2015. doi: 10.1109/CISDA.2015.7208625
Abstract: Neuromorphic computing offers an attractive means for processing and learning complex real-world data. With the emergence of the memristor, the physical realization of cost-effective artificial neural networks is becoming viable, due to reduced area and increased performance metrics than strictly CMOS implementations. In the work presented here, memristors are utilized as synapses in the realization of a multi-purpose heterogeneous 3D neuromorphic fabric. This paper details our in-house memristor and 3D technologies in the design of a fabric that can perform real-world signal processing (i.e., image/video etc.) as well as everyday Boolean logic applications. The applicability of this fabric is therefore diverse with applications ranging from general-purpose and high performance logic computing to power-conservative image detection for mobile and defense applications. The proposed system is an area-effective heterogeneous 3D integration of memristive neural networks, that consumes significantly less power and allows for high speeds (3D ultra-high bandwidth connectivity) in comparison to a purely CMOS 2D implementation. Images and results provided will illustrate our state of the art 3D and memristor technology capabilities for the realization of the proposed 3D memristive neural fabric. Simulation results also show the results for mapping Boolean logic functions and images onto perceptron based neural networks. Results demonstrate the proof of concept of this system, which is the first step in the physical realization of the multi-purpose heterogeneous 3D memristive neuromorphic fabric.
keywords: Boolean functions;CMOS integrated circuits;fabrics;memristors;neural chips;perceptrons;signal processing;three-dimensional integrated circuits;3D memristive neural fabric;3D technology;Boolean logic function application;CMOS implementation;area effective heterogeneous 3D integration;artificial neural network;complementary metal oxide semiconductor;defense application;extendable multipurpose 3D neuromorphic fabric;logic computing;memristive neural network;mobile application;nanoscale memristor;neuromorphic computing;perceptron;power conservative image detection;signal processing;Decision support systems;Fabrics;Memristors;Metals;Neuromorphics;Neurons;Three-dimensional displays;3D integrated circuits;Neuromorphics;image processing;memristor;nanoelectronics;neural networks (ID#: 16-9263)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7208625&isnumber=7208613
Tankasala, S.P.; Doynov, P., "Multi scale multi directional shear operator for personal recognition using Conjunctival vasculature," in Technologies for Homeland Security (HST), 2015 IEEE International Symposium on, pp. 1-6, 14-16 April 2015. doi: 10.1109/THS.2015.7225292
Abstract: In this paper, we present the results of a study on utilization of Conjunctival vasculature pattern as a biometric modality for personal identification. The visible red blood vessel patterns on the sclera of the eye is gaining acceptance as a biometric modality due to its proven uniqueness and easy accessibility for imaging in the visible spectrum. After acquisition, the images of Conjunctival vascular patterns are enhanced using the difference of Gaussian (DoG). The feature extraction is performed using a multi-scale, multi-directional shear operator (Shearlet transform). Linear discriminant analysis (LDA), neural networks (NN) and pairwise distance metrics were used for classification. In the study, images of 50 subjects are acquired with a DSLR camera at different gazes and multiple distances (CIBIT-I dataset). Additionally, the performance of the proposed algorithms is tested using different gaze images acquired from 35 subjects using an iPhone (CIBIT-II dataset). ROC AUC analysis is used to test the classification performance. Areas under the curve (AUC) and equal error rates (EER) are reported for all acquisition scenarios and different processing algorithms. The best EER value of 0.29% is obtained for a CIBIT-I dataset using NN and a 2.44% EER value for a CIBIT-II dataset using LDA.
keywords: Gaussian processes;biometrics (access control);blood vessels;error statistics;eye;feature extraction;image classification;image segmentation;neural nets;transforms;CIBIT-I dataset;CIBIT-II dataset;DSLR camera;DoG;EER value;LDA;ROC AUC analysis;Shearlet transform;areas under the curve;biometric modality;classification performance;conjunctival vasculature pattern;difference of Gaussian;equal error rate;eye;feature extraction;gaze images;iPhone;image classification;linear discriminant analysis;multiscale multidirectional shear operator;neural network;pairwise distance metric;personal identification;personal recognition;red blood vessel pattern;sclera;visible spectrum;Artificial neural networks;Cameras;Feature extraction;Image segmentation;Measurement;Transforms;Biometrics;Conjunctival vasculature;Difference of Gaussian;Linear discriminant analysis;Neural Networks;Ocular biometrics;Shearlet transforms (ID#: 16-9264)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7225292&isnumber=7190491
Adenusi, D.; Kuboye, B.M.; Alese, B.K.; Thompson, A.F.-B., "Development of cyber situation awareness model," in Cyber Situational Awareness, Data Analytics and Assessment (CyberSA), 2015 International Conference on, pp. 1-11, 8-9 June 2015. doi: 10.1109/CyberSA.2015.7166135
Abstract: This study designed and simulated cyber situation awareness model for gaining experience of cyberspace condition. This was with a view to timely detecting anomalous activities and taking proactive decision safeguard the cyberspace. The situation awareness model was modelled using Artificial Intelligence (AI) technique. The cyber situation perception sub-model of the situation awareness model was modelled using Artificial Neural Networks (ANN). The comprehension and projection submodels of the situation awareness model were modelled using Rule-Based Reasoning (RBR) techniques. The cyber situation perception sub-model was simulated in MATLAB 7.0 using standard intrusion dataset of KDD'99. The cyber situation perception sub-model was evaluated for threats detection accuracy using precision, recall and overall accuracy metrics. The simulation result obtained for the performance metrics showed that the cyber-situation sub-model of the cybersituation model better with increase in number of training data records. The cyber situation model designed was able to meet its overall goal of assisting network administrators to gain experience of cyberspace condition. The model was capable of sensing the cyberspace condition, perform analysis based on the sensed condition and predicting the near future condition of the cyberspace.
keywords: artificial intelligence;inference mechanisms;knowledge based systems;mathematics computing;neural nets;security of data;AI technique;ANN;Matlab 7.0;RBR techniques;anomalous activities detection;artificial intelligence;artificial neural networks;cyber situation awareness model;cyberspace condition;proactive decision safeguard;rule-based reasoning;training data records;Artificial neural networks;Computational modeling;Computer security;Cyberspace;Data models;Intrusion detection;Mathematical model;Artificial Intelligence;Awareness;cyber-situation;cybersecurity;cyberspace 9265) 9265)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7166135&isnumber=7166109
Seemakurthi, P.; Shuhao Zhang; Yibing Qi, "Detection of fraudulent financial reports with machine learning techniques," in Systems and Information Engineering Design Symposium (SIEDS), 2015, pp. 358-361, 24-24 April 2015. doi: 10.1109/SIEDS.2015.7117005
Abstract: This paper describes our efforts to apply various advanced supervised machine learning and natural language processing techniques, including Binomial Logistic Regression, Support Vector Machines, Neural Networks, Ensemble Techniques, and Latent Dirichlet Allocation (LDA), to the problem of detecting fraud in financial reporting documents available from the United States' Security and Exchange Commission EDGAR database. Specifically, we apply LDA to a collection of type 10-K financial reports and to generate document-topic frequency matrix, and then submit these data to a series of advanced classification algorithms. We then apply evaluation metrics, such as Precision, Receiver Operating Characteristic Curve, and Area Under the Curve to evaluate the performance of each algorithm. We conclude that these methods show promise and suggest applying the approach to a larger set of input documents.
keywords: document handling;financial data processing;fraud;learning (artificial intelligence);matrix algebra;natural language processing;neural nets;pattern classification;regression analysis;security of data;support vector machines;EDGAR database;LDA;Security and Exchange Commission;United States;area under the curve;binomial logistic regression;classification algorithms;document-topic frequency matrix;ensemble techniques;evaluation metrics;financial reporting documents;fraudulent financial reports detection;latent Dirichlet allocation;natural language processing techniques;neural networks;precision;receiver operating characteristic curve;supervised machine learning techniques;support vector machines;Accuracy;Classification algorithms;Correlation;Logistics;Natural language processing;Neural networks;Support vector machines;Ensemble;Financial Fraud Detection;Latent Dirichlet Allocation;Machine Learning;Natural Language Processing;Support Vector Machines (ID#: 16-9266)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7117005&isnumber=7116953
Khalifa, A.A.; Hassan, M.A.; Khalid, T.A.; Hamdoun, H., "Comparison between mixed binary classification and voting technique for active user authentication using mouse dynamics," in Computing, Control, Networking, Electronics and Embedded Systems Engineering (ICCNEEE), 2015 International Conference on, pp. 281-286, 7-9 Sept. 2015. doi: 10.1109/ICCNEEE.2015.7381378
Abstract: The rapid proliferation of computing processing power has facilitated a rise in the adoption of computers in various aspects of human lives. From education to shopping and other everyday activities to critical applications in finance, banking and, recently, degree awarding online education. Several approaches for user authentication based on Behavioral Biometrics (BB) were suggested in order to identify unique signature/footprint for improved matching accuracy for genuine users and flagging for abnormal behaviors from intruders. In this paper we present a comparison between two classification algorithms for identifying users' behavior using mouse dynamics. The algorithms are based on support vector machines (SVM) classifier allowing for direct comparison between different authentication-based metrics. The voting technique shows low False Acceptance Rate(FAR) and noticeably small learning time; making it more suitable for incorporation within different authentication applications.
keywords: behavioural sciences computing;government data processing;learning (artificial intelligence);mouse controllers (computers);pattern classification;security of data;support vector machines;FAR;SVM;active user authentication;behavioral biometrics;false acceptance rate;learning time;mixed binary classification;mouse dynamics;support vector machine;voting technique;Artificial neural networks;Biometrics (access control);active authentication;machine learning;mouse dynamics;pattern recognition;support vector machines (ID#: 16-9267)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7381378&isnumber=7381351
Parker, G.G.; Weaver, W.W.; Robinett, R.D.; Wilson, D.G., "Optimal DC microgrid power apportionment and closed loop storage control to mitigate source and load transients," in Resilience Week (RWS), 2015, pp. 1-7, 18-20 Aug. 2015. doi: 10.1109/RWEEK.2015.7287420
Abstract: This paper considers the optimal management of an N source, DC microgrid with time-varying sources and loads. Optimality is defined as minimizing the amount of power lost through the boost converters that connect the N sources to the DC bus. The optimal power apportionment strategy is part of an overall grid management system that also includes control laws that manage bus voltage and boost currents using distributed energy storage. The performance of an optimal power apportionment strategy is compared to an existing, alternate approach using a three source simulation for both source and load step transients. The optimal power apportionment strategy is shown to use less power while maintaining bus voltage in the presence of both load and source transients. Since the optimal solution requires information exchange between all the sources, there is an opportunity for malicious attack. The ability of the strategy to maintain the desired bus voltage in the presence of an uncommunicated source failure is also presented.
keywords: closed loop systems;distributed power generation;electric current control;failure analysis;load regulation;optimal control;power control;power distribution control;power distribution faults;power system security;voltage control;bus voltage management;closed loop storage control;current boosting;distributed energy storage;load transient mitigation;malicious attack;optimal DC microgrid power apportionment;overall grid management system;power lost amount minimization;source transient mitigation;uncommunicated source failure;Computational modeling;Energy storage;Feedforward neural networks;Microgrids;Steady-state;Transient analysis;Voltage control (ID#: 16-9268)
URL: http://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=7287420&isnumber=7287407
Note:
Articles listed on these pages have been found on publicly available internet pages and are cited with links to those pages. Some of the information included herein has been reprinted with permission from the authors or data repositories. Direct any requests via Email to news@scienceofsecurity.net for removal of the links or modifications to specific citations. Please include the ID# of the specific citation in your correspondence.