International Journal of Intelligent Systems Technologies and Applications (51 papers in press)
Multi-Agent System Approach for Improve Real-Time Visual Summaries of Geographical Data Streams
by Zina Bouattou, Robert Laurini, Hafida Belbachir
Abstract: Interdisciplinary efforts are needed to support visual exploration and analysis of spatio-temporal data streams from sensor networks, depending on the size and complexity of data that must be analysed before being displayed. For this, several emerging approaches have been proposed known as Visual Summaries of a set of data that will help users find what is most important and interesting to visualize in the mass of available information. In this paper, we present an approach for generating automatically the visual summaries in real time. For this approach, we have adopted chorem-based visual representations of territories issued from both geometric and semantic generalizations. Roger Brunet defined Chorems as schematized visual representations of territories. Our approach relies on a multi-agent framework; an extraction knowledge agent is able to extract important spatiotemporal patterns of data streams coming from a sensor network as interesting regions, and a visualization agent which displays those patterns as simplified maps. The novelty is the multi-agent semantic generalization and its combination with cartographic generalization to generate maps on-the-fly. We validate this model with an example taken from the meteorology field.
Keywords: Geovisualization; sensors networks; important patterns; chorems; multi-agent systems (MAS); visual summaries; automatic generalization.
Automatic Arabic Text Summarization System (AATSS) Based on Morphological Analysis
by Laiali Almazaydeh
Abstract: This Automatic text summarization is a challenging task. Existing techniques suffer from inadequate investigation for Arabic text. The majority of Arabic words are derived according to the morphology that is more complicated than English and other Natural languages. The approach reported in this paper elaborates a morphological analysis to develop a new Automatic Arabic text summarization. The proposed approach is based on three main phases: the text tokenization phase, the trilateral root extraction phase and the aggregate similarity computation phase. The proposed approach has been extensively experimented on samples from the benchmark datasets of EASC. Experimental results show the ability of the proposed system to extract relevant sentences that reflects the summary of Arabic text, as well as a comparable performance with other reported techniques in the literature.
Keywords: Information retrieval; Arabic language; summarization; trilateral rootroot; aggregate similarity.
A Two-Phase Human Activity Classification Design Using Accelerometer Data from Smartphone
by Syed Arsalan Ali, Rooh Ul Amin
Abstract: This paper presents a two phase design for the classification of the human movement activities of Running, Walking and Standing by using accelerometer data logged from smart phone. The first phase of the design is, the two-stage filter applied for the process of raw acceleration data collected from the accelerometer; and the second phase of the design is, the classification of human movements using Logistic Regression with Gradient Descent Algorithm as classifier from the clean data produced by first phase of the design. The raw acceleration data from the tri-axial accelerometer of smart phone contains three axes accelerations. This acceleration data is collected separately for each movement activity and is used for the training and testing of design by using a simulation environment. The activity classification results are promising and show that the proposed design provides an accuracy of more than 96 percent on both train data and the separate independent test data for the classification of human movement activities of Running, Walking and Standing.
Keywords: Pattern recognition; Smartphone accelerometer; Two-Stage filter; Gradient Descent Algorithm; Classifier.
An Online English-Khmer hybrid Machine Translation System
by Suraiya Jabin, Niladri Chatterjee, Suos Samak, Kim Sokphyrum, Javier Sola
Abstract: The present paper describes design of an online hybrid Machine Translation (MT) system involving a low-resource language Khmer. A Statistical Machine Translation system is typically trained on a large parallel corpus in order to compute the translation probabilities, and on an even larger quantities of monolingual data to compute the target language n-gram probabilities that are used to design the language model. The proposed system uses the open source SMT toolkit DoMY CE for training corpus (English-Khmer parallel corpus and also developing the corpus) combined with a post processing step of using parts of speech tagger to further enhance the quality of target language sentence. Web technologies, such as, Python, Apache2 (Web server), HTML, XML, and XSLT for developing an online translation system. Language Model, Translation Model and Decoder Configurations are done in accordance with recent literature using the DOMY toolkit. The parallel corpora have been prepared from various sources, such as ICT books for high school students, translated texts from chemist, education glossaries, law and history articles, country and city names, and the Headley Khmer-English dictionary. Experimental results are presented to demonstrate the success of phrase based SMT methods where English has been used as source and Khmer, a low-resource language as a target language. We used NIST (National Institute of Standards and Technology) and BLEU (BiLingual Evaluation Understudy) metrics have been used evaluate the performance of the system based on the training corpus built. In our experiments the proposed model achieved significantly good BLEU and NIST scores. Khmer being the official language of Cambodia the demand for such an automated system has been on the rise for several decades. The present work is one of the useful efforts towards meeting this demand.
Keywords: Computational Linguistic; English-Khmer Parallel Corpora; English-Khmer translation; Khmer language; Moses toolkit; Statistical Machine Translation; Phrase-based model; Rule-based machine translation system; Hybrid Machine Translation system.
Robust Feature Selection Algorithm based on Transductive SVM Wrapper and Genetic Algorithm: Application on Computer-Aided Glaucoma Classification
by Nawel Zemmal, Mokhtar Sellami, Djamel Zenakhra, Nilanjan Dey, Amira S. Ashour
Abstract: Glaucoma has become a devastating disease after cataract to cause blindness. Thus, early diagnoses for glaucoma can prevent the vision loss. Computer Aided Diagnosis (CAD) systems, which automate the process of ocular disease detection, are urgently needed to alleviate the burden on the clinicians. In the current work, advanced machine learning algorithms are investigated to propose a robust system for Glaucoma diagnosis based on retinal images. Three features extraction methods, namely the Gray-Level Co-Occurrence Matrix (GLCM), Hu moments and Central moments are combined to form the entry feature vector. To select the most relevant features and taking into account at the same time the unlabeled data existing in medical databases, a new scheme of feature selection algorithm is proposed. It is based on Transductive SVM Wrapper and Genetic algorithm, which automatically detect and classify the glaucoma disease using fundus images. The effectiveness of the proposed GA-TSVM is evaluated on a public retinal database RIM-ONE using the classification accuracy, sensitivity, and specificity metrics. The experimental results established that with 16% of labeled data, the proposed system could easily distinguish between the normal and the affected glaucoma cases.
Keywords: Glaucoma disease; Computer Aided Diagnosis; Genetic Algorithm; Feature Extraction and Selection; Semi supervised Learning; Transductive Support Vector Machine (TSVM).
Enhanced Particle Swarm Optimization Algorithms for Multiple-Input Multiple-Output System Modelling using Convolved Gaussian Process Models
by Gang Cao, Edmund Lai, Fakhrul Alam
Abstract: Convolved Gaussian Process (CGP) is able to capture the correlations not only between inputs and outputs but also among the outputs. This allows a superior performance of using CGP than standard Gaussian Process (GP) in the modelling of Multiple-Input Multiple-Output (MIMO) systems when observations are missing for some of outputs. Similar to standard GP, a key issue of CGP is the learning of hyperparameters from a set of input-output observations. It typically performed by maximizing the Log-Likelihood (LL) function which leads to an unconstrained nonlinear and non-convex optimization problem. Algorithms such as Conjugate Gradient (CG) or Broyden-Fletcher-Goldfarb-Shanno (BFGS) are commonly used but they often get stuck in local optima, especially for CGP where there are more hyperparameters. In addition, the LL value is not a reliable indicator for judging the quality intermediate models in the optimization process. In this paper, we propose to use enhanced Particle Swarm Optimization (PSO) algorithms to solve this problem by minimizing the model output error instead. This optimization criterion enables the quality of intermediate solutions to be directly observable during the optimization process. Two enhancements to the standard PSO algorithm which make use of gradient information and the multi-start technique are proposed. Simulation results on the modelling of both linear and nonlinear systems demonstrate the effectiveness of minimizing the model output error to learn hyperparameters and the performance of using enhanced algorithms.
Keywords: Enhanced PSO; Convolved Gaussian Process Models; Hyperparameters Learning.
Support Vector Machine based Fault Detection and Diagnosis for HVAC Systems
by Jiaming Li
Abstract: Various faults occurred in the Heating, Ventilation and Air-Conditioning (HVAC) systems usually lead to more energy consumption and worse thermal comfort inevitably. This paper presents a feasible and valid solution of HVAC fault detection and diagnosis problem based on statistical machine learning technology. It learns the consistent nature of different types of faults of HVAC operation based on Support Vector Machine (SVM), and then identify types of fault in all subsystems using the statistical relationships between groups of measurements. In order to speed up the learning process, Principle Component Analysis (PCA) has been applied to compress the training data. Our approach models the dynamical sub-systems and sequence data in HVAC system. The learnt models can then be used for automatic fault detection and diagnosis. The approach has been tested on commercial HVAC systems. It had successfully detected and identified a number of typical AHU faults.
Keywords: Fault detection and diagnosis; FDD; Machine learning; SVM; HVAC system; Principle Component Analysis; PCA.
Efficient Blind Nonparametric Dependent Signal Extraction Algorithm for Determined and Underdetermined Mixtures
by Fasong WANG
Abstract: Blind extraction or separation statistically independent source signals from linear mixtures have been well studied in the last two decades by searching for local extrema of certain objective functions, such as non-Gaussianity (NG) measure. Blind source extraction (BSE) algorithm for extracting statistically dependent source signals from underdetermined and determined linear mixtures is derived using nonparametric NG measure in this paper. After showing that maximization of the NG measure can also separate or extract the statistically weak dependent source signals, the nonparametric NG measure is defined by statistical distances between different distributions of separated signals based on cumulative density function (CDF) instead of traditional probability density function (PDF), which can be estimated by the quantiles and order statistics (OS) using the norm efficiently. The nonparametric NG measure is optimized by a deflation procedure to extract or separate the dependent source signals. Simulation results for synthesis and real world data show that the proposed nonparametric extraction algorithm can extract the desired dependent source signals and yield ideal performance.
Keywords: blind source separation; non-Gaussianity measure; independent component analysis; probability density function; dependent component analysis; underdetermined blind source extraction.
Indic Script Identification from Handwritten Document Images
by Pawan Kumar Singh, Ram Sarkar, Mita Nasipuri
Abstract: Script identification plays an important role in document image processing especially for multilingual environment. This paper hires two conventional textural methods for the recognition of the scripts of the handwritten documents inscribed in different Indic scripts. The first method extracts the well-known Haralick features from Spatial Gray-Level Dependence Matrix (SGLDM) and the second method computes the fractal dimension by using Segmentation-Based Fractal Texture Analysis (SFTA). Finally, a 104-element feature vector is constructed from the features designed by these two methods. The proposed technique is then evaluated on a total dataset comprising of 360 handwritten document pages written in 12 Indian official scripts namely, Bangla, Devanagari, Gujarati, Gurumukhi, Kannada, Malayalam, Manipuri, Oriya, Tamil, Telugu, Urdu and Roman. Experimentations using multiple classifiers reveal that Multi Layer Perceptron (MLP) shows the highest identification accuracy of 96.94%. Encouraging outcome confirms the efficacy of customary textural features to handwritten Indic script identification.
Keywords: Script Identification; Handwritten Indic documents; Textural Features; Spatial Gray-Level Dependence Matrix; Segmentation-Based Fractal Texture Analysis; Statistical Significance Tests.
Combining RSS-SVM with Genetic Algorithm for Arabic Opinions Analysis
by Amel ZIANI, Nabiha Azizi, Djamel Zenakhra, Soraya Cheriguene
Abstract: The Arabic language has drawn the attention of researchers in data mining due to its large-scale users, but it presents challenges because of its richness and complex morphology. That explains the growing importance of the Arabic sentiment analysis and precisely the Arabic opinion detection and classification areas. Thus, the most accurate classification technique used in this area which proven by several previous works is the Support Vector Machine classifier (SVM). This last, is able to increase the rates in opinion mining but with use of very small number of features. Hence, reducing features vector can alternate the system performance by deleting some pertinent ones. To overcome these two constraints (features vector size and considerate all extracted primitives), our idea is to use Random Sub Space algorithm (RSS) at first stage to generate several features vectors with limited size; and to replace the decision tree base classifier of RSS with more accurate classifier which is the SVM. Therefore, each one of the generated features subsets will be the entry of an individual SVM classifier. Despite the obtained high results of this proposed approach, another proposition was implemented in order to enhance the previous algorithm by using the genetic algorithm as subset features generator based on correlation criteria to eliminate the random choice used by RSS and to prevent the use of incoherent features subsets. For that, features extraction stage is considered the main step in opinion mining process. Wherefore, to extract and to calculate the best informative linguistic and statistic features, we struggled with the lack of labeled Arabic dataset. So, a translated SentiWordNet dataset has been used to overcome this problem and to ameliorate the proposed system performance. The experiments results applied on one thousand (1000) labeled reviews collected from Arabic Algerian newspapers and manually annotated are very promising.
Keywords: Arabic opinion mining; SentiWordNet; Machine learning; SVM (Support Vector Machine); RSS (Random Sub Space); GA (Genetic Algorithm).
Facial beauty analysis by age and gender
by Manal El Rhazi, Arsalane Zarghili, Aicha Majda, Anissa Bouzalmat, Ayat Allah Oufkir
Abstract: The face is the first source of information that inspires the attractiveness of a human being, for this reason; several studies were conducted in the aesthetic medicine and the image processing to analyze the aesthetic quality of an adult human face.
This paper proposes an automatic procedure for the analysis of facial beauty. First, we detect the face zone on an image and its features areas, then we present our novel method to extract features corners, and finally we analyze the facial aesthetic quality.
Experimental results show that our method can extract the features corners accurately for the majority of faces presented in the ECVP and FEI images databases, and that there exist a difference in the facial beauty analysis by gender and age, due to anatomic differences in specific facial areas between the categories.
Keywords: Facial features; Plastic surgery; Facial attractiveness; Facial beauty analysis; Image processing; Image analysis.
Enhancement based Background Separation Techniques for Fruit Grading and Sorting
by Jasmeen Gill
Abstract: Image processing plays a remarkable role in the automation of fruit grading and sorting. While grading the fruit, accurate extraction of fruit object from the image (background separation) is the chief concern. For extraction of fruit, appropriate segmentation technique is employed; and to accomplish it accurately, enhancement must be performed prior to segmentation. However, majority of the researchers emphasized over fruit segmentation alone. This communication is intended to show the potential of enhancement techniques when combined with fruit image segmentation. Besides, it presents a comparative analysis of enhancement based background separation techniques for fruit grading and sorting. For this purpose, four main techniques, namely, Contrast limited adaptive histogram equalization (CLAHE) method, Gaussian filter, Median filter and Wiener filter were utilized for enhancement and Basic Global thresholding, Adaptive thresholding, Otsu thresholding and Otsu-HSV thresholding were applied for segmentation. 16 sub-models were developed by combining each enhancement method with every segmentation technique. Afterwards, the image quality of the sub-models was validated using quantitative as well as qualitative analyses. Test results demonstrate that CLAHE/Otsu-HSV model outperformed the others for fruit grading and sorting.
Keywords: Digital image processing; Segmentation; Enhancement; Fruit grading and sorting; Background separation; Otsu-HSV segmentation.
Soft Neural Network based Block Chain Risk Estimation
by Ganglong Duan, Wenxiu Hu, Yu Tian
Abstract: Financial risk refers to the uncertainty caused by the change of the economic and financial conditions. As a kind of economic phenomenon, the financial risk is objective and can not be eliminated. At present, there are still some imperfect aspects in the research of financial risk assessment. In order to achieve the purpose of comprehensive evaluation of financial risks, the paper analyses the methodology of soft computing and neural networks. The basic function of financial risk monitoring and evaluation system is to forecast the trend of financial activities and risk status, and this is also the fundamental function and objectives of the assessment system. We use BP neural network theory to establish the logistics finance risk evaluation model, using BP neural network structure and training principles to train sample data. The soft computing method is based on the factors of uncertainty and irrationality, which breaks through the limitation of traditional hard computing. There is a consistency between the fuzzy thinking principle of soft computing method and the attribute and structure of the objective world, therefore, soft computing can be used in field of financial risk assessment.
Keywords: Block chain; financial risk; assessment; neural network; soft computing.
Data Flow Tracking based Block Chain Modelling
by Ganglong Duan, Wenxiu Hu, Yu Tian
Abstract: This article carries on the analysis to the recent global financial industry focus block chain, introduces the concept and scope of the technology, and to the financial payment service as an example, analyzes the block chain technology to the center, to trust four characteristics, collective maintenance and safety database etc.. In this area chain based technology in the financial field of application and Research on the current situation, the international payment as an example, analyzes the block chain payment mode and the traditional mode of payment differences, and that the international financial industry attention block chain technology, nature is to build a flat global integrated settlement system. Finally, we summarize the trend of the development of block chain technology innovation, and put forward some key issues that should be paid attention to. Block chain is a kind of computing paradigm to the center of the infrastructure and distributed with new bitcoin digital encryption currency popularity gradually rise, at present it has attracted great attention and concern of government departments, financial institutions, enterprises and capital market. The block chain technology has decentralized, time series data, the collective maintenance, programmable and safe and reliable characteristics, especially suitable for construction of the monetary system, the financial system and macro social system programming. The rapid development of the financial chain block technology applications such as digital currency, smart contracts cause new financial risk, which brings a series of challenges to the existing financial supervision system in China.
Keywords: data stream; tracking algorithm; finance; block chain; technical framework.
AN EFFICIENT MEDICAL IMAGE WATERMARKING TECHNIQUE USING INTEGER WAVELET TRANSFORM AND QUICK/FAST RESPONSE CODES
by K.J. Kavitha
Abstract: Securing the medical images, to make it tamper free is a terribly difficult task. This challenge is with efficiency handled with the assistance, digital watermarking techniques. With the assistance of this growing technology we are able to evaluate validation, dependability, privacy and integrity of the medical images. Several algorithms are enforced on this technology. The Digital Watermarking (DWM) is implemented in two main domains: transform & spatial. The DWM is mostly implemented using the transform techniques such as Singular Valued Decomposition (SVD), Discrete Cosine Transform (DCT), Discrete Wavelet Transform (DWT), combination of DCT and DWT and also with the combination of DWT and SVD. These days the work is extended with Integer Wavelet Transform (IWT). One of the foremost challenges in these technologies is information embedding capability or we say; number of bits hidden in the cover information. This parameter is considered for evaluation of the system since as we know that the more number of bits; the distortion of the medical information is also more and vice versa. The distortion of the information is strictly avoided within the case of medical and military applications. To beat this downside of the technology, these days research work goes on to implement digital watermarking technique with less information embedding capability. One of the possible ways to reduce the number of bits in information is to use Quick /fast response code (QR Code). The QR code consumes a less space compared to the other existing formats available such as barcode. In this paper Associate in nursing approach is planned to implement the digital watermarking technique for the medical images that includes the following techniques: Integer wavelets Transform (IWT), Bit plane methodology and QR code. For the proposed system, the watermarked image is evaluated against a number of the parameters to grasp the potency of technique being utilized. The experiment is dispensed for 2 totally different bit planes and also the results are compared to point out embedding within which of the bit plane range ends up in additional potency and eventually the conclusion is formed.
Keywords: Watermarking; DWT; QR Code.
Special Issue on: Soft Computing Approaches and Intelligent Systems
A comprehensive solution for Risk Management in software development projects
by Raghavi Bhujang, Suma V
Abstract: Risk Management is a never ending process in the life cycle of a software project. For efficient risk management, it is necessary to have deep understanding of risk, its types, mitigation strategies so that either the risk is avoidable in future or the impact of the same can be brought down. This paper essentially deals with comprehensive understanding of risk from different perspectives and offers a solution- research methodology for efficient risk management. Risk is prioritized based on the impact which is measured with the help of essential parameters that are inevitable to complete the development phase of the project. Based on the priority, solution is applied either to mitigate or to bring down the impact of these risks oriented towards the successful completion of the software development project on time.
Keywords: Risk Management; Classification of Risk; Risk Mitigation Strategies; Risk Management solution-methodology.
Area Efficient SDR Receiver Without and With Dynamic Partial Reconfiguration
by T. THAMMI REDDY, Madhavi B.K, K. Lal Kishore
Abstract: The Xilinx offered Dynamic Partial Reconfiguration (DPR) is a very efficient mechanism to utilize FPGA resources for realizing several application categories. In the context of realizing signal processing data path in area efficient applications several researchers attempted to DPR. The present work demonstrates a generic framework for implementing Software Defined Radio (SDR) based communication system through DPR technique. The work switches contexts with two partial reconfiguration blocks. The first of the DPR block is Spectrum Estimation and second one is FSK receiver. The Spectrum Estimation module is implemented through streaming type FFT architecture. The non-coherent FSK demodulator is implemented using frequency shifting and filtering stages. The major blocks are streaming FFT, complex multiplier, Numerically Controlled Oscillator (NCO), Finite Impulse Response (FIR) filter, peak detection logic and energy detection logic. The completely developed FSK receiver is simulated using Modelsim simulator and results are verified for functional correctness. Xilinx Zynq 7010 SoC with partial reconfiguration feature is used to develop this application. A signal generator that generates FSK signal with symbol rate 64 Kbps bit rate is used to drive the ADC input for testing the application. The work demonstrates novel direction for SDR implementation with DPR technique towards low area FPGA implementation. The results shows that the chosen system can be implementable in a low cost FPGA with 45% lesser DSP48 slice utilization compared to without DPR case. Reduced power dissipation is observed as a by-product of DPR.
Keywords: Dynamic Partial Reconfiguration (DPR); FSK generic demodulator; Xilinx Zynq 7010 SoC; DDS; FIR filter; Redpitaya.
A new method for the optical flow estimation and segmentation of moving objects "NMES"
by Amina Ourchani
Abstract: Segmention of moving objects in video sequence is a hard, but essential task in a large number of applications in computer vision . Most existing methods gave accepted results only in the case where both object and background are rigid, because of serious occlusions and complex computation, which presents limitations in case of occlusions and shadows. This paper focuses on developing a new method for discriminate moving objects from a static background, focusing on the combination of motion, color and texture features. First, we have used block-matching which is one of the most powerful methods for computing the optical flow, we also have taken in consideration the result of frame difference, to improve the quality of the optical flow. Moreover, we have used the k-means clustering algorithm owing to group the pixels, having similar motion, color and texture features. Second, the result of the grouping pixels is used as an input in Chan-Vese model, in order to attract the evolving contour of moving objects contours. To evaluate the performance of ourvproposed methode, we experiment it on challenging sequences. It has shown that our method provides an improved segmentation results.
Keywords: Segmentation; moving object; optical flow; color feature; texture feature; k-means; frame difference; Chan-Vese model; block-matching; occlusion.
Visual substitution system for room labels identification based on text detection and recognition
by Hanen Jabnoun
Abstract: Vision is an essential human sense which is playing the most important role in human perception about surrounding environment. However such information is generally inaccessible for blind people. In order to enable these persons overcome such incapability, we propose a system for identifying the neighbouring scene. In fact, the environment is labelled as a first step and then we present a personalized and adaptive method for labels detection and recognition.
The automatic analysis involves the detection of personalized markers employed to extract the region of interest followed by the perspectives rectification based on homography estimation. We use the stroke width transform to extract the text from the detected label and crop it into words and letters. The extracted text is identified using images correlation. Finally, as a substitution for visual modality, a vocal transformation of the recognized text is performed.
Keywords: Image rectification; text detection; homography estimation; stroke width transform; images correlation; blind people.
Ordered Weighted Averaging Operator used to Enhance Accuracy of Fuzzy Predictor based on Genetic Algorithm
by Rohit Garg, Md. Tabrez Nafis, Bindu Garg
Abstract: Accuracy is one of the most important aspects while designing forecasting model. Particularly in fuzzy prediction system, accuracy considerably depends on subjectively decided parameters such as membership function and relative weight of past observations. In this regard, we have proposed a novel concept to optimize ordered weighted aggregation (OWA) based fuzzy time series predictor (FTSP) using genetic algorithm (GA). Firstly, accurateness of FTSP is enhanced by applying effective method of aggregation on past observations using OWA weights. These weights are determined on the basis of importance of fuzzy set in the system by employing regularly increasing monotonic (RIM) quantifiers. Subsequently, GA is used to optimize membership functions of FTSP by generating its wide range of parameters in the region of time series. Lastly, this model is capable of controlling its performance by varying GA parameters. To assess proposed method, we used dataset of enrollments and outpatient visits, as used by almost all previous research in this domain. Evaluation results indicate coalescing OWA and GA for FTSP significantly reduced mean square error (MSE) and average forecasting error rate (AFER).
Keywords: Time Series analysis; Fuzzy Logic; Genetic algorithm; Optimization; Ordered Weighted Aggregation.
MACHINE LEARNING TECHNIQUES USING PYTHON FOR DATA ANALYSIS IN PERFORMANCE EVALUATION
by J. V. N. Lakshmi
Abstract: Machine learning algorithms are used to parallelize the workloads. To achieve paramount performance required parameters are tuned in the algorithms. The jobs are implemented using machine learning techniques using various parameters. The performance is examined by executing various features and verifying time constraints depending on the assignments for a cluster. The attempt is made to obtain minimum execution time using python language for implementing machine learning algorithms. Supervised and unsupervised techniques of machine learning algorithms are used differentiating the performance evaluation and time efficiency. Linear regression, logistic Regression and K- means clustering techniques are used to evaluate the Data Analytic jobs. This implementation reveals the best performance of supervised algorithms over unsupervised for data analysis. This paper is an attempt made to analyze the machine learning techniques and evaluate the timer feature on various methods irrespective of supervised or unsupervised.
Keywords: Machine Learning; Linear Regression; Logistic regression; K- Means; EM.
Wind energy potential estimation with prediction of wind speed distribution
by MANISH KUMAR, Cherian Samuel
Abstract: Integration of renewable distributed generation technology with radial power distribution system is rapidly growing. Amongst the number of available renewable energy sources, the role of wind energy in the power sector is very important. In this work, authors have estimated the wind energy potential at Banaras Hindu University (BHU), India. Estimation of wind energy potential depends on statistical analysis of wind speed distribution, which will give us the best-fitted probability density function of the desired location. For statistical analysis of wind speed distribution, authors have taken hourly data of BHU centre from the source of National Renewable Energy Laboratory (NREL), USA for the twelve-month period. A statistical R programming language tool has been used for the computational work. In this work authors have considered Weibull, Gamma and Lognormal Probability distributions for getting best-fitted distribution. After the analysis, result shows for the given wind speed data of BHU area, Weibull distribution is the best-fitted one.
Keywords: Statistical Analysis; Wind Speed; Wind Energy Potential; Renewable Energy; Renewable Distributed Generation; Probability Distribution; Goodness-of-Fit.
Enhanced Portable Text to Speech Converter for visually impaired
by Chithra Selvaraj
Abstract: The Handwritten Text Reader is designed to help the visually impaired listen to an audio read-back of printed and handwritten scanned text. A hand-held page scanner is used to scan the text to be read. The image from the scanner is sent to the application in the paired Android phone over Bluetooth. Tesseract OCR engine is used to extract the text from the image, and this extracted text is converted to speech. Tesseract OCR engine is further trained to recognize handwritten text for a specific user. This OCR engine is trained with handwritten datasets. In addition to English, the application supports two regional languages - Hindi and Bengali.
Keywords: Handwriting Recognition; Tesseract; Optical Character Recognition; Text to Speech; Android; visually impaired;.
Multi spectral image classification using cluster ensemble technique
by Radhika K, Varadarajan S, Muralimohanbabu Y
Abstract: Satellite image classification is an imperative system utilized as a part of remote sensing. Primary data of extraordinary significance to different difficulties can be acquired straightforwardly from Land-cover observation. It is necessary to discuss the issue of supervised classification of satellite images in the view of cluster ensemble. Different information partitions inferred by various clustering methods can be gathered into a new solution by cluster ensembles. supervised iterative Expectation-Maximization method can be initialized by cluster ensemble based strategy which will be examined in the paper. The upgraded parameter set obtained from the expectationmaximization (EM) step is trained by maximum likely-hood classifier to classify the rest of the pixels. The performance of clustering of the proposed method is compared with individual clustering of the ensemble for medium resolution and a very high spatial resolution images. The accuracy measurements have done with different test points. The state of art techniques are giving less accuracy and are not well defined. This paper will explore the possibility of all accuracy parameters with supervised classification results. The accuracy parameters are well tested and compared in this paper with various start of art techniques. The approach has been tested on different ground truth points for validation to get better results.
Keywords: Classification; Image; Satellite; Ensemble; Resolution.
Simultaneous Scheduling of Jobs, Machines and Tools Considering Tool Transfer Times in Multi Machine FMS Using New Nature Inspired Algorithms
by Sivarami Reddy Narapureddy, Ramamurthy D.V., Prahlada Rao K.
Abstract: This article addresses simultaneous scheduling of jobs, machines and tools considering tool transfer times between machines, to generate best optimal sequences that minimize makespan in a multi-machine Flexible Manufacturing System. Performance of FMS is expected to improve by effective utilization of its resources, by proper integration and synchronization of their scheduling. The aim of this article is to address joint scheduling of jobs, machine and tools in a FMS consisting of machines, central tool magazine and tool transporter considering tool transfer times. Three nature inspired algorithms namely Symbiotic Organisms Search algorithm, Crow search algorithm and Flower pollination algorithm, have been proposed for solving joint jobs, machine and tool scheduling problems considering tool transfer times between machines with minimum makespan as objective. The proposed algorithms are numerically tested on various problems and the results are compared .The results show that FPA algorithm yields better results for simultaneous scheduling of jobs, machines and tools considering tool transfer times between machines.
Keywords: Flexible manufacturing systems; Simultaneous Scheduling of jobs; tools and machines; Tool Transporter; Symbiotic organisms search algorithm; Crow search algorithm; Flower pollination algorithm.
Grammar Rule based Sentiment categorization model for classification of Tamil Tweets
by Nadana Ravishankar T., Shriram R.
Abstract: Movies have always been a part of life of majority of individuals across the globe. The advent of social networking websites have enabled people to easily and publicly express their ideas on a movie/product in such a way that it reaches millions of people within no time. The aim of this research is to implement a tool that would be helpful in predicting the genre of the movies as perceived by the audience through linguistic rules and Natural Language Processing Tool Kit. This paper especially focuses on development of rule based classification algorithms for the domain of Tamil movies. We develop a sentiment categorizing tool for Tamil tweets using grammar rules and a tool has been developed using Python and NLP Tool Kit. Further, we have designed a model to determine the opinion in addition to genre classification of Tamil movies. For this work we select a set of genres of Tamil movies with public tweets based on sentiment analysis. We find that the developed tool classifies the genre of a particular movie by taking into consideration of Tamil tweets by the users. We empirically validate our approach with domain experts and baseline models.
Keywords: Tamil Movie Tweets; Sentiment analysis; NLP; Data mining.
Empirical Study of Feature Selection Methods over Classification Algorithms
by Bhalaji Natarajan, Sundharakumar KB, Chithra S
Abstract: Feature selection methods are deployed in machine learning algorithms for reducing the redundancy in the data set and to increase the clarity in the system models without loss of much information. The objective of this article is to investigate the performance of feature selection methods when they are exposed to different data sets and different classification algorithms. In this article, we have investigated standard parameters such as accuracy, precision, and recall over two feature selection algorithms namely Chi Square feature selection and Boruta feature selection algorithms. Observations of the experiments conducted using R studio resulted around 5-6% increased performance in above said parameters when they were exposed to Boruta feature selection algorithm. The experiment was done one two different datasets with a different set of features and we have used the following five standard classification algorithms - Naive Bayes, Decision Tree, Support Vector Machines, Random Forest and Gradient Boosting.
Keywords: Classification;Feature Selection;Boruta;Chi-Square;Ensemble classifiers.
Improvement of Power quality in Microgrids using Predictive Controller
by Prajith Prabhakar
Abstract: Microgrid is a low voltage grid which is subjected to disturbances. The distributed energy resources are connected to the LV bus by means of power electronic converters. The power quality issues are analyzed with Model Predictive control (MPC) algorithm in both- Grid connected & Islanded modes. But the Islanded mode is analyzed separately. A new model is suggested for the islanded mode with non- linear loads. Predictive control based Distribution static compensator (DSTATCOM) injects voltages at the point of common coupling (PCC) in the Microgrid. Reactive power compensation and mitigation of harmonics are done with the help of new proposed model. The principles and the proposed algorithm are verified by MATLAB/SIMULINK simulation. Simulation results suggest that this control method can decrease the power quality issues and increase the microgrid stability.
Keywords: Power Quality; Active power control; microgrid; predictive control; DSTATCOM;.
GA tuned Two Degree of Freedom PID Controller for Time Delay Systems
by FEBINA BEEVI PARAYI, T.K. Sunil Kumar, Jeevamma Jacob
Abstract: In this study, the authors address the problem of design of a Two Degree Of Freedom (2-DOF) PID controller for time delay systems. The proposed approach is to design 2-DOF PID controller using a novel method which combines model order reduction, approximate model matching concepts as well as optimization techniques. The conventional PID controllers are usually in 1-DOF structure. The problem of finding the parameters of the 2-DOF PID controller is formulated as that of obtaining the solution of a set of non-homogeneous linear equations. The proposed method not only ensures the stability of the closed loop system with a 2-DOF PID controller but also satisfies the required performance criteria. The developed method does not pose any restriction on either the order of the model or on the structure/order of the controller transfer function. Moreover, this method is computationally simple and easy to implement. Simulation results demonstrate the effectiveness of the proposed method.
Keywords: Model Matching; Model Order Reduction; Optimization Techniques; Time Delay; 2-DOF PID Controller.
Reactive Frequency Band based Real -Time Motor Imagery Classification
by Sumanta Bhattacharyya, Manoj Kumar Mukul
Abstract: The requirement of an effective online processing algorithm becomes very vital to fulfilling the demand of the low- cost brain-computer interface (BCI) system. Based on the variation of relative spectral power, the authors proposed an exclusive method of signal processing for the real-time classification of EEG data related with the imagination of left-hand and right-hand movement. The proposed algorithm contains reactive frequency band (RFB) identification and feature extraction process, which make it very first and robust unsupervised machine learning algorithm for the real-time BCI. The RFB has been identified by applying short time Fourier transforms (STFT) based dominant frequency detection algorithm over the training data set. Based on the identified RFB, the temporal relative spectral power (TRSP) based feature extraction process has been applied to the testing dataset. The estimated "feature" further classified as per probabilistic Bayesian classifier. The effectiveness of the proposed RFB detection method of EEG signal is validated by self-generated artificial sine wave signal, single subject and nine subject movement imagery (MI) BCI competition dataset. The proposed method of EEG signal processing outperformed the conventional wavelet-based BCI competition II results and the wavelet-based algorithm applied over the BCI competition IV dataset.
Keywords: Brain Computer Interface (BCI); Electroencephalogram (EEG); Movement Imagery; Reactive Frequency Band (RFB); Short Time Fourier Transforms (STFT); Temporal Relative Spectral Power (TRSP); Wavelet.
Special Issue on: Inventive Systems and Internet of Things
Mixed Integer Programming for Vehicle Routing Problem with Time Windows
by Divya Aggarwal, Vijay Kumar
Abstract: Being a key element in logistics distribution, Vehicle Routing Problem becomes an importance research topic in management and computation science. Vehicle Routing Problem with time windows is a specialization of Vehicle Routing Problem. In this paper, a brief description of vehicle routing problem is presented. A Mixed Integer Programming is utilized to solve the vehicle routing problem with time windows. A novel mathematical model of MIP is formulated and implemented using IBM CPLEX. A novel constraint is designed to optimize the number of vehicle used. The proposed model is used to optimize both transportation cost and number of vehicle used simultaneously. The proposed model is tested on two well-known instances of Solomons benchmark test problem. Experimental results illustrate that the proposed formulation provides promising solutions in reasonable computation time. The sensitivity analysis of customer nodes is also studied.
Keywords: Vehicle Routing Problem; Time Windows; Solomon’s Instance; CPLEX; MIP.
EFFICIENT MULTIMEDIA CONTENT STORAGE AND ALLOCATION IN MULTIDIMENSIONAL CLOUD COMPUTING RESOURCES
by Sivaraman Eswaran, Manickachezian R
Abstract: Optimal management of the cloud resources for multimedia contents is the important aim of this research. In our previous work, Multiple Kernel Learning with Support Vector Machine (MKL-SVM) is introduced which can achieve a balanced resource usage with multimedia user request. However existing work do not concentrate on caching mechanism which might lead to more computational overhead. To solve this problem, new method is proposed namely Improved Storage and Scheduling of Multimedia Contents in Cloud Storage (ISS-MCCS). In this work, Fuzzy Neural Network Classification (FNNC) is utilized for handling the server clusters with unevenness. Then task scheduling is done using Hybrid Genetic-Cuckoo Search Algorithm (HGCSA) where Hybrid fuzzy weighting scheme is used for the fitness evaluation. Finally Adaptive Replacement Cache (ARS) is integrated to optimize memory. The overall assessment of the research work is done in cloudsim environment which proves it can manage the multimedia contents with efficiently.
Keywords: Multimedia contents; multiple QoS; optimized scheduling; efficient load balancing; adaptive replacement; fuzzy neural network classification.
Context Aware Reliable Sensor Selection in IoT
by K. R. Remesh Babu Raman, M. Vishnu Prathap, Philip Samuel
Abstract: Internet of Things (IoT) is a computing concept where physical objects with embedded sensors connect to the Internet and can identify themselves to other devices. Today the number of devices connected to the Internet is increasing rapidly and they want to communicate each other for different purposes. The future Internet will comprise of billions of intelligent communicating objects having capabilities for sensing, actuating, and data processing. Each object in this Cyber Physical Systems (CPS) will have one or more embedded sensors that will capture huge amount of data. Managing these data in cloud and obtaining the relevant data from appropriate sensors are important concerns. For information retrieval context awareness is important. Usually users need information from these sensors depending upon several factors like location, accuracy level, etc. The proposed method senses reliable data from a sensor environment that satisfies user contexts. It also contains different functionalities like user addition, location sensing, context specification, user context counting, and selection of current best results.
Keywords: Cloud Computing; Context Awareness; IoT; Sensor selection; Cloud of Things; Resource management.
Lagrangian Relaxation for Distribution Networks with Cross-Docking Center
by Manpreet Singh Bhangu, Rimmi Anand, Vijay Kumar
Abstract: This paper proposes a cross-docking in logistics network that aims to reduce the transportation costs. The proposed strategy eliminates need of inventory for storing commodities. It consists of two main stages. During first echelon, the number and location of cross-dock warehouses are determined. In second echelon, the allocation of warehouse as cross-dock to distribution centers is determined. Each warehouse has limited capacity and distribution center can supplied by only one cross-dock. In this paper, the problem is mathematical formulated using a mixed-integer programming model. This formulation is used the concept of cross-docking allocation and commodities distribution. The Lagrangian relaxation approach is proposed to solve logistics network problem. Experimental results reveal that the proposed approach provides optimal solution in a reasonable time.
Keywords: Facility location; Network design; Mixed-integer programming; Lagrangian relaxation; Merge-in-Transit.
Factors Influencing Regression Testing on Cloud and On-Premises: An Analysis
by Suma V, Narasimha Murthy M S
Abstract: Since, the evolution of software, it has laid its impact in all domains of operations wherein software industry has taken up the major role. Hence, it is required to upgrade software as the market demands so that it can sustain in the industrial environment. However, one of the most important criteria for software industries to survive is development of high quality software which can completely satisfy customers. In order to achieve the above mentioned goal, it becomes mandatory for the software organizations to adopt themselves to the market dynamics. Since, cloud computing has not taken wide popularity as one of the promising technology in the current situation, IT services have marched themselves to serve the needs of the society through cloud based technology. Testing applications using cloud has further become one of the dominating areas of operations. The main objective of this paper is therefore to analyse the effectiveness, and efficiency of testing applications in cloud environment. This paper further put forth a case study which involves an empirical investigation carried out in leading software company follows cloud technology in their day to day flow of developmental activities. A comprehensive analysis is conducted upon sampled data which is collected in two different domains namely health care and telecom domain of the company under investigation. From the analysis, it is observed that testing applications in cloud model is a good practice against conventional mode of software testing (on-premises). The results also depict testing applications in cloud environment improves the performance of various parameters of testing process. Further, this inference paves way to carry out further research to formulate effective strategies to test applications in cloud.
Keywords: Software Engineering; Software Testing; Cloud Computing; Regression Testing; Software Quality; Total Customer Satisfaction.
A Hybrid Test Prioritization Technique for Combinatorial Testing
by Preeti Satish, Krishnan Rangarajan
Abstract: IoT systems comprise of multiple devices connected together, to perform an intelligent task in real time. Such systems have to be meticulously tested in order to avoid hazards situations. Combinatorial testing technique can effectively test such complex IoT systems with reduced effort as it generates fewer test cases with adequate coverage. It basically tests the interactions that exist between values of different parameters and practically faults upto six-way interactions have been found successfully. Prioritization of combinatorial tests deals with finding an ideal order of the test cases so that faults are detected early. Recent approaches to prioritization problem are either coverage based or parameter-value weight based for two-way or three-way interaction strengths separately. In this paper, we present a hybrid prioritization technique for combinatorial testing that combines both weight based and interaction coverage based approaches. We derive a combined weight to each test case considering user given weights denoting the importance of values of parameters and the interaction coverage up to six-way interactions. To the best of our knowledge, no research has been carried out yet, that accounts for higher order combinations up to six-way at a time. To demonstrate the effectiveness of our algorithm, we have conducted initial synthetic experiments on various covering arrays, and measured the effectiveness with t-Rate of fault of detection metric. The results are promising in covering the combinations early.
Keywords: IoT systems; combinatorial testing; prioritization; combinatorial coverage; weight based; hybrid technique; interaction testing; interaction strength.
Intelligent systems for Redundancy Removal with Proficient Run Length Coding and statistical analysis using regression
by V.R. PRAKASH, S. Nagarajan
Abstract: The surveillance video aspect has been one of the key technologies in various tactical monitoring. However, the quantum of analysis with proper implication of video quality subjected to enormous amount of time might degrade its error metrics. So in order to analyse this quantum has been made with the hierarchical order wherein four videos where taken and its peak errors where being analysed. The significance of the work is dealt with feature extraction and then comparison with input and extracted texture followed by feature analysis with cosine angle distance. Finally, a multiple regression analysis has been developed with PSNR as dependant variable where video size and execution time are taken as independent variable. The significance of regression has been based on prediction equation has been done in order to near optimality of PSNR value for varying video size and execution time.
Keywords: Proficient Run Length Coding; Regression analysis.
An Intelligent Inventive System for Personalized Web Page Recommendation based on Ontology Semantics
by Gerard Deepak, Ansaf Ahmed, Skanda B
Abstract: Owing to the information diversity in the Web and its dynamically changing contents, extraction of relevant information from the web is a huge challenge. With the World Wide Web transforming into a more organized Semantic Web, the incorporation of Semantic techniques to retrieve relevant information is highly necessary. In this paper, a dynamic ontology alignment technique for recommending relevant Web Pages is proposed. The strategy focuses on knowledge tree construction by computing the semantic similarity between the query terms as well as the ontological entities. Furthermore, the semantic similarity is again computed between nodes of the constructed knowledge tree and URLs in the URL repository to recommend relevant Web Pages. The dynamic ontology alignment by computing their respective semantic similarity constitutes Ontology Semantics. Personalization is achieved by prioritization of Web Pages by Content Based Analysis of the users Web Usage Data. An overall accuracy of 87.73 % is achieved by the proposed approach.
Keywords: Ontologies; Personalized; Semantic Strategy; Web Page Recommendation System; Web Search.
Input Data Transition Aware Adaptive Frequency Scaling based Energy Efficient MAC Architecture for Low Power DSP Applications
by Haripriya D, Govindaraju C, Sumathi M
Abstract: In this paper, input transition aware adaptive frequency scaling based low power 8 bit Multiplier-Accumulator (MAC) architecture for Digital Signal Processing (DSP) has been presented. As the data transition causes dynamic power dissipation in the electronic circuits, it becomes mandatory to minimize the data transition to minimize the dynamic power. A novel adaptive frequency scaling scheme based on the number of input data transitions is implemented and the dynamic power has been minimized to a greater extent. The proposed input data transition aware adaptive frequency scaling is very effective method to minimize the dynamic power consumption without degrading the performance of the system. The input data transition detector circuit in the proposed low power MAC detects the transition and applies the adaptive frequency scaling based on the number of transitions, so that a notable amount of dynamic power is reduced by the proposed scheme. The dynamic power consumed by the conventional MAC is 77.05mW when all inputs are switching and it is only 2.370mW for the proposed MAC with the same simulation parameters. The proposed MAC consumes 96.92% less power than the conventional MAC for the same set of inputs and simulation environment at 27
Keywords: Adaptive Frequency Scaling; Digital Signal Processing; Dynamic Power; Energy Efficient; Low Power; Multiply and Accumulate; Transition Aware.
SURVEY ON TECHNIQUES OF FAULT DETECTION- ROOKIES VANTAGE POINT
by Vijaya Bharathi Manjeti, Koteswara Rao Kodepogu
Abstract: Cutting edge frameworks are turning out to be exceedingly configurable to fulfil the shifting needs of clients and clients. Programming product offerings are consequently turning into a typical pattern in programming improvement to decrease cost by empowering deliberate, expansive scale reuse. A few shortcomings may be uncovered just if a specific mix of components is chosen in the conveyed items. Yet, testing all mixes is typically not possible by and by, because of their to a great degree extensive numbers. Combinatorial testing is a method to produce littler test suites for which all mixes of t elements are ensured to be tried. In this paper, we display a few hypotheses depicting the likelihood of irregular testing to recognize connection blames and contrast the outcomes with combinatorial testing. For instance, an irregular testing turns out to be considerably more viable as the quantity of elements increments and focalizes toward equivalent adequacy with combinatorial testing. Be that as it may, when imperatives are available among elements, then irregular testing can passage subjectively more awful than combinatorial testing. Subsequently, with a specific end goal to have a reasonable effect, future research ought to concentrate on combinatorial testing.
Keywords: fault; framework; vantage.
Special Issue on: Computational Intelligence informatics and Information Security
Image Preprocessing of Icing Transmission Line based on Fuzzy Clustering
by Wang Jing, Zhang Hangbo, Han Ming, Yang Zhengyan
Abstract: Propose a kind of electric transmission lines icing image segmentation algorithm based on quick bi-circulating level set to improve effect of electric transmission lines icing image segmentation algorithm. Firstly, carry out pretreatment to graying of electric transmission lines icing image and decrease noise influence in image; secondly, improve fuzzy c mean value clustering algorithm and propose SPKFCM algorithm by increasing space penalty function, which is used for automatic initialization of quick bi-circulating level set algorithm; finally, it is shown that proposed algorithm has better segmentation effect and segmentation efficiency by experiment contrast on electric transmission lines icing image.
Keywords: Quick bi-circulation; Level set; Electric transmission lines; Icing image; Segmentation.
A Study on Power Balance Control Strategies of Mining Variable Speed Magnetic Coupling Based on Fuzzy Self-adaptive PID
by Wang Lei, Jia Zhen Yuan, Zhang Li, Wang Xin Ming, Li Liang
Abstract: This paper studied fuzzy self-adaptive PID control strategies that are applicable to mining variable speed magnetic coupling based on analysis on the magnetic coupling mathematical model fitted by a large number of data simulation and calculation by combining traditional PID control strategies and fuzzy control methods and then carried out experiments for further verification. It can be proved by experimental analysis that such control strategy can not only solve multi-motor driving power balance problems effectively, but adapt to working condition changes and revise control parameters automatically, which can provide theoretical support for project practices.
Keywords: magnetic coupling; PID control; fuzzy control; power balance.
Exploration on Predicting Breast Cancer Stage with the Aid of Redesigned ANN Incorporated with Enhanced Social Spider Optimization (ESSO) Technique
by Ramani Selvanambi, Jaisankar Natarajan
Abstract: The core intention of this work is to predict the breast cancer stage as benignant or malignant from the given input dataset with parameters such as instance Clump Thickness, Uniformity of Cell Size, Uniformity of Cell Shape, Marginal Adhesion, Single Epithelial Cell Size, Bare Nuclei, Bland Chromatin, Normal Nucleoli and Mitoses. Predicting the cancer stage helps to determine the best way to contain and eliminate the breast cancer. In this process various techniques are used, one of the classification methods used is Artificial Neural Network (ANN) which it is trained with several training algorithms and the selected training algorithm is Levenberg-Marquardt. This training algorithm performs in the better way and gives minimum error value. In order to obtain the better prediction the default structure of ANN is redesigned using optimization techniques. In the default structure, one hidden layer of the network includes ten neurons. To improve the structural design (hidden layer and neuron) diverse optimization techniques are used for example Cuckoo Search (CS), Particle Swarm Optimization (PSO), Social Spider Optimization (SSO) and Enhanced Social Spider Optimization (ESSO). Our results shows the Enhanced Social Spider Optimization (ESSO) is employed better and evaluates the metrics as Accuracy 97%, Sensitivity 98%, and Specificity 95% compared with other techniques. The accuracy is fine tuned in this work were contrasted with existing work and the stage of breast cancer is predicted.
Keywords: Artificial Neural Network(ANN); Breast cancer; Levenberg-Marquardt (LM) algorithm; Feed Forward Back Propagation(FFBN); Enhanced Social Spider Optimization(ESSO) algorithm.
Prediction of oil Production based on SVM Optimized Multi Objective Particle Swarm Optimization
by Rong Wang-Yin, Rui Zhou, An Xian-Shao, Jing Yu-Pang
Abstract: The unstable availability and capacity exist in energy efficiency optimization regarding to renewable energy supply for cellular network due to several factors, such as climate change, high oil resource consumption and energy safety etc. The energy supply sustainability problem is modeled as optimization solution problem aiming at maximizing NP-hard network energy residue ratio (ERR). The network ERR maximization algorithm has been put forward through analysis of relationship between energy efficiency and energy depletion rate (EDR). This algorithm is worked by adopting cross-layer manner, which first maximizes the link energy efficiency by the power control at physical (PHY) layer and then maximizes network ERR by the access control at media access control (MAC). The simulation results show that the network ERR maximization algorithm performs excellently in improving network lifetime as well as increasing the number of users served by renewable energy.
Keywords: energy efficiency; cellular network; renewable energy; energy residue rate; energy depletion rate.
An Improved Adaptation Algorithm for Signer-independent Sign Language Recognition
by Wang Min, Wang Ya, Zhu Xiao-Juan
Abstract: Sign language recognition is a technology that can present sign language in a understandable form, in order to achieve barrier-free communication between deaf and normal. In order to solve the differences in sign language data issues and the lack of training samples of manpower caused by low recognition of non-specific language, presents MLLRMAP adaptive progressive non-specific integrated manpower language recognition framework. This approach optimizes the division MLLR regression class to provide more accurate initial MAP model, which give full play to the rapidity and the MAP MLLR progressive. Then introduced MCE model parameter estimation algorithm to compensate for the limitations of the model parameters adaptive method to further reduce the system error rate and accelerate the recognition speed. Meanwhile, for the MCE algorithm computationally intensive problems proposed improvements. Experimental results show that the adaptive sign language data required for this algorithm is less than traditional MLLR and MAP methods, while improved average recognition rate by 15.6%.
Keywords: machine learning; signer-independent sign language recognition; mllr algorithm; map algorithm; model parameter estimate.
Design of Remote Control System for Intelligent Irrigation based on Zigbee and GPRS
by Zhang Hongbo
Abstract: The sensor nodes of WSNs (Wireless Sensor Networks)-based agricultural monitoring system are required to collect the environmental data periodically, including temperature and humidity, and additionally the any time monitoring shall be required to be realized among the monitored area with minimized quantity of sensor nodes. To meet the mentioned goal, a SPMSC (Solar Power-Moving Schedule of mobile -based optimal Coverage) is proposed. In accordance with the content of SPMSC, every sensor node is able to obtain energy through the solar panel. The moving schedule of sensor nodes is determined through predicting the quantity of solar power as to realize the optimal coverage project with the minimized quantity of sensor nodes and the energy consumption. As the simulation result indicates, the proposed SPMSC, compared with other similar projects, has the 4% decrease of sensor node quantity and the 10% increase of the network lifetime.
Keywords: WSNs (Wireless Sensor Networks); Agricultural monitoring; Intelligence algorithm; Coverage; Network lifetime; Irrigation system.
Modification of AES using Genetic Algorithms for High-Definition Image Encryption
by Sayantani Basu, Marimuthu K, Rajkumar S, Niranchana R
Abstract: Genetic Algorithms (GAs) have proved to be a powerful tool in cryptography. The standard AES algorithm incurs high computation costs and has problems in pattern formation when applied to image encryption. It is also prone to the differential and side channel attacks. In this paper, a modification of AES has been proposed for High-Definition (HD) Image Encryption that uses a genetic approach for the S-Box Generation and Key Expansion processes. S-Boxes have been evolved to have high nonlinearity and low transparency. Also, in the key generation, the process has been randomized by using only genetic operators. As a result of applying genetic modifications to AES, the proposed algorithm has been found to be more secure against the differential and side-channel attacks and shows reduced pattern appearance when applied for encryption of HD images.
Keywords: AES; GA; image encryption; s-box; key generation.
A Study On Flow Based Classification Models Using Machine Learning Techniques
by Chokkanathan Kothandapani
Abstract: Network traffic management facing a lot ofchallenges in the recent years because of the continuous and fast development in network scale, number of consumers and type of known and unknown applications over the network. Traffic Classification is a key factor for providing the Quality of network Services (QoS) and also playing a vital role in handling the delay and congestion during the network transmission. In spite of many algorithms existing in the field of network traffic classification, Machine learning algorithms are playing key role. In this paper I would like to discuss about flow based classification models such as port based, payload based, statistical based and behavioral based classifications which are frequently used for identifying traffic classes. Techniques (Methods): Widely used techniques such as port based, pay-load based, behavioral based and statistical based classification models are discussed and sample data sets have been produced with graphical notations to strengthen the analysis process. Findings: This analysis provides the capability to distinguish and isolate the traffic flows into real-time and non-real-time traffic flow according to the process category. Since we are experiencing the integrated environment of multiple fields its very difficult to classify the traffic. So the comprehensive study on these classification models will provide the clarity on how the traffic is classified with different attribute sets and to understand the quality of service and issues of each classification model. Applications:In a wide-range of network environment, traffic classification is more important to provide the differentiated service quality based on the various types of applications and also for monitoring the security over the traffic. In a network traffic analysis, network classification is the first step and core element for network intrusion detection system.There are many network servicing areas where we need to identify and classify the trafficsuch as routing, firewall access-control, policy based routing, traffic billing need to be differentiated and their quality of service has to be assured.
Keywords: Network Classifications Models; Network Traffic Analysis; Quality of Service; Intrusion Detection System.
A hybrid approach to improve the quality of software fault prediction using Na
by Sathyaraj Ragupathi, Prabu S
Abstract: This paper considers an improvisation in software fault prediction research area using supervised classification algorithms and it mainly focuses to increase the performance of fault prediction. In this paper, we propose a hybrid prediction model using na
Keywords: Naïve Bayes; K-NN; software fault prediction; classification; ensemble.
Labeled Decision Making Method based on Neural Network Model and Pruning Algorithm
by Jian-hai Du
Abstract: Aiming at the advantages in commonly used method of generating decision tree, a method of generating cost-sensitive decision tree based on the correlation degree of neural network attributes is proposed through quoting the correlation degree of neural network attributes and cost-sensitive learning. This method reduces the condition attributes by using rough set theory, and takes the correlation degree and cost performance of attributes as the bases of split node to build the cost-sensitive decision tree by using modified information gain method during the process of building decision tree. It is shown in test result that such method is superior to commonly used method of generating decision tree in classification accuracy and the number of nodes generated.
Keywords: neural network; rough set; cost-sensitive; labeled tree reduction; decision-making.
Design of Water Quality Monitoring System based on WSN and ZigBee
by Zhang Yuan, Qi Lan
Abstract: In the Thesis, based on actual demand, a set of water quality monitoring system on the basis of wireless sensor network, ZigBee technology and GPRS technology is designed to realize a large scale of real-time water quality monitoring. Wireless sensor network is applied to city macrozone water quality monitoring system, in the meantime, wireless transmission technologies such as ZigBee and GPRS are combined to realize collection and transmission of water quality data, and technologies such as sleep mode management, automatic alarm of exceeding water area, self-adaptively high-speed collection, preferred transmission and network cluster node topology control optimization are adopted to improve energy utilization efficiency of the system, which reduces the cost of system maintenance and use and meets with actual application demands. System networking and deployment are convenient comparatively with great expandability and application prospects.
Keywords: Water quality monitoring; Wireless sensor network; ZigBee; GPRS; Data transmission.