Showing papers in "International Journal of Intelligent Systems and Applications in 2018"

Blockchain with Internet of Things: Benefits, Challenges, and Future Directions

Abstract: The Internet of Things (IoT) has extended the internet connectivity to reach not just computers and humans, but most of our environment things. The IoT has the potential to connect billions of objects simultaneously which has the impact of improving information sharing needs that result in improving our life. Although the IoT benefits are unlimited, there are many challenges facing adopting the IoT in the real world due to its centralized server/client model. For instance, scalability and security issues that arise due to the excessive numbers of IoT objects in the network. The server/client model requires all devices to be connected and authenticated through the server, which creates a single point of failure. Therefore, moving the IoT system into the decentralized path may be the right decision. One of the popular decentralization systems is blockchain. The Blockchain is a powerful technology that decentralizes computation and management processes which can solve many of IoT issues, especially security. This paper provides an overview of the integration of the blockchain with the IoT with highlighting the integration benefits and challenges. The future research directions of blockchain with IoT are also discussed. We conclude that the combination of blockchain and IoT can provide a powerful approach which can significantly pave the way for new business models and distributed applications.

Development and Analysis of Artificial Neural Network Models for Rainfall Prediction by Using Time-Series Data

Abstract: Time Series data is large in volume, highly dimensional and continuous updating. Time series data analysis for forecasting, is one of the most important aspects of the practical usage. Accurate rainfall forecasting with the help of time series data analysis will help in evaluating drought and flooding situations in advance. In this paper, Artificial Neural Network (ANN) technique has been used to develop one-month and twomonth ahead forecasting models for rainfall prediction using monthly rainfall data of Northern India. In these model, Feed Forward Neural Network (FFNN) using Back Propagation Algorithm and LevenbergMarquardt training function has been used. The performance of both the models has been assessed based on Regression Analysis, Mean Square Error (MSE) and Magnitude of Relative Error (MRE). Proposed ANN model showed optimistic results for both the models for forecasting and found one month ahead forecasting model perform better than two months ahead forecasting model. This paper also gives some future directions for rainfall prediction and time series data analysis research.

Development and Implementation of the Technical Accident Prevention Subsystem for the Smart Home System

Abstract: The structure of the technical accident prevention subsystem for the smart home system has been developed in the article. The subsystem model based on Petri network, model based on neural network and physical model using the Arduino microcontroller have been realized in the development process. The subsystem research results with the use of the developed models, softand hardware tools are also presented.

Time Dependence of the Output Signal Morphology for Nonlinear Oscillator Neuron Based on Van der Pol Model

Abstract: Time-frequency and time dependence of the output signal morphology of nonlinear oscillator neuron based on Van der Pol model using analytical and numerical methods were investigated. Threshold effect neuron, when it is exposed to external non-stationary signals that vary in shape, frequency and amplitude was considered.

Fuzzy Logic Based Energy Aware Routing Protocol for Internet of Things

Abstract: Maximizing the network lifetime is one of the major challenges in Low Power and Lossy Networks (LLN). Routing plays a major role in LLN, for minimizing the energy consumption across the network nodes. IPv6 Routing Protocol for Low Power and Lossy Networks (RPL) is a standardized routing protocol for LLN. Though, RPL fulfilled the necessity of LLN, several issues like increasing the energy efficiency, quality of service and the network lifetime are to be focused. In LNN, the inefficient route selection results in increased network traffic, energy depletion and packet loss ratio across the network. In this paper, we propose a fuzzy logic based energy aware routing protocol (FLEARPL), which considers the routing metrics load, residual energy (RER) and expected transmission count (ETX) for the best route selection. FLEA-RPL applies fuzzy logic over these metrics, to select the best route to transfer the network data efficiently. The COOJA simulator is used to assess the efficiency of the proposed FLEA-RPL. The FLEA-RPL protocol is compared with similar protocol standard RPL, MRHOF (ETX) based RPL (MRHOFRPL) and FL-RPL. The simulation result shows that FLEA-RPL improves the network lifetime by 10-12% and packet delivery ratio by 2-5%.

The Information Support of Virtual Research Teams by Means of Cloud Managers

Abstract: The article deals with the creation of virtual research teams of scientists from various geographically distributed organizations united for joint interdisciplinary researches. Library social institutions are the satellites of virtual research teams and have to implement information and communication support of scientific researches. The use of cloud managers by academic libraries is proposed as platforms to facilitate remote collaborative work of the participants of the virtual research teams. The research of number of free cloud managers and their capabilities was held. The most successful cloud manager for supporting the scientific work of virtual research teams was selected by using hierarchy analysis method.

A New Robust and Imperceptible Image Watermarking Scheme Based on Hybrid Transform and PSO

Abstract: In this paper, a new robust and imperceptible digital image watermarking scheme that can overcome the limitation of traditional wavelet-based image watermarking schemes is proposed using hybrid transforms viz. Lifting wavelet transform (LWT), discrete cosine transform (DCT) and singular value decomposition (SVD). The scheme uses canny edge detector to select blocks with higher edge pixels. Two reference sub-images, which are used as the point of reference for watermark embedding and extraction, have been formed from selected blocks based on the number of edges. To achieve a better trade-off between imperceptibility and robustness, multiple scaling factors (MSF) have been employed to modulate different ranges of singular value coefficients during watermark embedding process. Particle swarm optimization (PSO) algorithm has been adopted to obtain optimized MSF. The performance of the proposed scheme has been assessed under different conditions and the experimental results, which are obtained from computer simulation, verifies that the proposed scheme achieves enhanced robustness against various attacks performed. Moreover, the performance of the proposed scheme is compared with the other existing schemes and the results of comparison confirm that our proposed scheme outperforms previous existing schemes in terms of robustness and imperceptibility.

Classification and Regression Trees (CART) for Predictive Modeling in Blended Learning

Abstract: Today, Internet and Web technologies not only provide students opportunities for flexible interactivity with study materials, peers and instructors, but also generate large amounts of usage data that can be processed and reveal behavioral patterns of study and learning. This study analyzed data extracted from a Moodle-based blended learning course, to build a student model that predicts course performance. CART decision tree algorithm was used to classify students and predict those at risk, based on the impact of four online activities: message exchanging, group wiki content creation, course files opening and online quiz taking. The overall percentage of correct classifications was about 99.1%, proving the model sensitive to identify very specific

Technology of Gene Expression Profiles Filtering Based on Wavelet Analysis

Abstract: The paper presents the technology of gene expression profiles filtering based on the wavelet analysis methods. A structural block-chart of the wavelet-filtering process, which involves concurrent calculation of Shannon entropy for both the filtered data and allocated noise component is proposed. Simulation of the waveletfiltering process was performed with the use of orthogonal and biorthogonal wavelets on different levels of wavelet decomposition and with the use of various values of the thresholding coefficient. Result of the simulation has allowed us to propose the technology to determine the optimal parameters of the wavelet filter based on complex analysis of the filtered data and allocated noise component.

Spiking neural network and bull genetic algorithm for active vibration control

Abstract: Systems with flexible structures display vibration as a characteristic property. However, when exposed to disturbing forces, then the component and/or structural nature of such systems are damaged. Therefore, this paper proposes two heuristics approaches to reduce the unwanted structural response delivered due to the external excitation; namely, bull genetic algorithm and spiking neural network. The bull genetic algorithm is based on a new selection property inherited from the bull concept. On the other hand, spiking neural network possess more than one synaptic terminal between each neural network layer and each synaptic terminal is modelled with a different period of delay. Extensive simulations have been conducted using simulated platform of a flexible beam vibration. To validate the proposed approaches, we performed a qualitative comparison with other related approaches such as traditional genetic algorithm, general regression neural network, bees algorithm, and adaptive neuro-fuzzy inference system. Based on the obtained results, it is found that the proposed approaches have outperformed other approaches, while bull genetic algorithm has a 5.2% performance improvement over spiking neural network.

An Approach for the Generation of Higher Order Mutants Using Genetic Algorithms

Abstract: Mutation testing is a structural testing technique in which the effectiveness of a test suite is measured by the suite ability to detect seeded faults. One fault is seeded into a copy of the program, called mutant, leading to a large number of mutants with a high cost of compiling and running the test suite against the mutants. Moreover, many of the mutants produce the same output as the original program (called equivalent mutants), such mutants need to be minimized to produce accurate results. Higher order mutation testing aims at solving these problems by allowing more than one fault to be seeded in the mutant. Recent work in higher order mutation show promising result in reducing the cost of mutation testing and increasing the approach effectiveness. In this paper, we present an approach for generating higher order mutants using a genetic algorithm. The aim of the proposed approach is to produce subtle and harder to kill mutants, and reduce the percentage of produced equivalent mutants. A Java tool has been developed, called HOMJava (Higher Order Mutation for Java), which implements the proposed approach. An experimental study was performed to evaluate the effectiveness of the proposed approach. The results show that the approach was able to produce subtle higher order mutants, the fitness of mutants improved by almost 99% compared with the first order mutants used in the experiment. The percentage of produced equivalent mutants was about 4%.

An Effectiveness Evaluation of Information Technology of Gene Expression Profiles Processing for Gene Networks Reconstruction

Abstract: The paper presents the research results concerning an effectiveness evaluation of information technology of gene expression profiles processing for purpose of gene regulatory networks reconstruction. The information technology is presented as a structural blockchart of step-by-step stages of the studied data processing. The DNA microchips of patients, who were investigated on different types of cancer, were used as experimental data. The optimal parameters of data processing algorithm at appropriate stage of this process implementation by quantity criteria of data processing quality were determined during simulation. Validation of the reconstructed gene networks was performed with the use of ROC-analysis by comparison of character of genes interconnection in both the basic network and networks reconstructed based on the obtained biclusters.

Selfie Sign Language Recognition with Convolutional Neural Networks

Abstract: Extraction of complex head and hand movements along with their constantly changing shapes for recognition of sign language is considered a difficult problem in computer vision. This paper proposes the recognition of Indian sign language gestures using a powerful artificial intelligence tool, convolutional neural networks (CNN). Selfie mode continuous sign language video is the capture method used in this work, where a hearing-impaired person can operate the Sign language recognition (SLR) mobile application independently. Due to non-availability of datasets on mobile selfie sign language, we initiated to create the dataset with five different subjects performing 200 signs in 5 different viewing angles under various background environments. Each sign occupied for 60 frames or images in a video. CNN training is performed with 3 different sample sizes, each consisting of multiple sets of subjects and viewing angles. The remaining 2 samples are used for testing the trained CNN. Different CNN architectures were designed and tested with our selfie sign language data to obtain better accuracy in recognition. We achieved 92.88 % recognition rate compared to other classifier models reported on the same dataset.

A Fuzzy-based Adaptive Energy Efficient Load Distribution Scheme in Ad-hoc Networks

Abstract: The traditional energy aware routing policies are not capable enough to keep up with dynamic properties of mobile ad-hoc network (e.g., mobility, quick topology changes, link-layer contentions etc.) and do not offer adequate performance in high congested situations. In past decades, authors have expressed their concerns towards smart routing paradigms concerning lesser energy consumption. However, many of these proposals are not able to offer significant performance concerning the quality of service. Consequently, the pattern of interest shifts towards cross-layer energy optimization schemes. These proposals did use of lower layers’ special information and provide significant performance enhancements. Still, many of the issues are associated with these proposals. Moreover, many of the proposals consider idle and sleep power consumption which too causes a considerable amount of energy consumption. Nevertheless, these methods require complex synchronization and efficient coordination which is too inefficient for extremely variable networks (MANETs). To address these issues, we propose an effective fuzzybased energy efficient load distribution scheme which takes care of energy consumption considering congestion as a parameter. In comparison with some of the existing energy aware routing strategies, proposed method offers substantial improvements in terms of total energy consumption, network lifetime, total number of dead nodes, and average throughput.

Winograd’s Inequality: Effectiveness for Efficient Training of Deep Neural Networks

Abstract: Matrix multiplication is widely used in a variety of applications and is often one of the core components of many scientific computations. This paper will examine three algorithms to compute the product of two matrices: the Naive, Strassen’s and Winograd’s algorithms. One of the main factors of determining the efficiency of an algorithm is the execution time factor, how much time the algorithm takes to accomplish its work. All the three algorithms will be implemented and the execution time will be calculated and we find that Winograd’s algorithm is the best and fast method experimentally for finding matrix multiplication. Deep Neural Networks are used for many applications. Training a Deep Neural Network is a time consuming process, especially when the number of hidden layers and nodes is large. The mechanism of Backpropagation Algorithm and Boltzmann Machine Algorithm for training a Deep Neural Network is revisited and considered how the sum of weighted input is computed. The process of computing the sum of product of weight and input matrices is carried out for several hundreds of thousands of epochs during the training of Deep Neural Network. We propose to modify Backpropagation Algorithm and Boltzmann Machine Algorithm by using fast Winograd’s algorithm. Finally, we find that the proposed methods reduce the long training time of Deep Neural Network than existing direct methods.

A Review on Impacts of Power Quality, Control and Optimization Strategies of Integration of Renewable Energy Based Microgrid Operation

Abstract: Due to the global demand for energy saving and reduction of greenhouse gas emissions, utilization of renewable energy sources have increased in electricity networks. The negative aspects of this technology are very complex and not well known which affect reliability and robustness of the grids. Microgrids based on renewable energy sources have gained significant popularity, due to the major benefits it has to offer for solving the increasing energy demand. Harmonic distortion in microgrids caused by the non-linear loads is an essential topic of study necessary for the better understanding of power quality impacts in microgrids. The various control techniques utilized to curtail the power quality impacts on micro grids are reviewed in this paper. Also, Optimization based control techniques utilized for power quality improvement in microgrids is discussed in this review.

Intelligent Training Algorithm for Artificial Neural Network EEG Classifications

Abstract: Artificial neural networks (ANN) have been widely used in classification. They are complicated networks due to the training algorithm used to fix their weights. To achieve better neural network performance, many evolutionary and meta-heuristic algorithms are used to optimize the network weights. The aim of this paper is to implement recently evolutionary algorithms for optimizing neural weights such as Grass Root Optimization (GRO), Artificial Bee Colony (ABC), Cuckoo Search Optimization (CSA) and Practical Swarm Optimization (PSO). This ANN was examined to classify three classes of EEG signals healthy subjects, subjects with interictal epilepsy seizure, and subjects with ictal epilepsy seizures. The above training algorithms are compared according to classification rate, training and testing mean square error, average time, and maximum iteration.

Weight Assignment Algorithms for Designing Fully Connected Neural Network

Abstract: Soft computing is used to solve the problems where input data is incomplete or imprecise. This paper demonstrate designing fully connected neural network system using four different weight calculation algorithms. Input data for weight calculation is constructed in the matrix format based on the pairwise comparison of input constraints. This comparison is performed using saaty’s method. This input matrix helps to build judgment between several individuals, forming a single judgment. Algorithm considered here are Geometric average mean, Linear algebra calculation, Successive matrix squaring method, and analytical hierarchical processing method. Based on the quality parameter of performance, it is observed that analytical hierarchical processing is the most promising mathematical method for finding appropriate weight. Analytical hierarchical processing works on structuration of the problem into sub problems, Hence it the most prominent method for weight calculation in fully connected NN.

New Delay-based Fast Retransmission Policy for CMT-SCTP

Abstract: Concurrent Multipath Transfer (CMT) uses multi-homing feature of Stream Control Transmission Protocol (SCTP) to transfer data concurrently over the multiple paths. CMT provides bandwidth aggregation, fault tolerance, and reliability in multipath data transfer. In multipath data transmission, each path has different delay and bandwidth. Therefore, destination receives unordered data which causes receiver buffer blocking and unwanted congestion window (cwnd) reduction. Both the problem degrades the CMT performance significantly. Thus, this paper proposes a new delay-based fast retransmission policy to adjust the transmission rate of each path according to path delay. Simulation results show that the proposed approach achieves better throughput, reduces the number of the timeout and improves the cwnd growth. The proposed approach improved throughput up to 16% in variable packet loss and 18% in variable network delay environment.

Towards an Incremental Recommendation of POIs for Mobile Tourists without Profiles

Abstract: Mobile tourism or m-tourism can assist and help tourists anywhere and anytime face the overload of information that may appear in their smartphones. Indeed, these mobile users find difficulties in the choice of points of interest (POIs) that may interest them during their discovery of a new environment (a city, a university campus ...). In order to reduce the number of POIs to visit, the recommendation systems (RS) represent a good solution to guide each tourist towards personalized paths close to his instantaneous location during his visit. In this article, we focus on (1) the detection of the spatiotemporal context of the tourist to filter the POIs and (2) the use of the previous notations of the places. These two criteria make it possible to integrate the evolutionary context of the visit in order to predict incrementally the most relevant transitions to be borrowed by the tourists without profile. These predictions are calculated using collaborative filtering algorithms that require the collection of traces of tourists to better refine the recommendation of POIs. In our software prototype, we have adapted the SLOPE ONE algorithm to our context of discovering the city of Chlef to avoid problems like data scarcity, cold start and scalability. In order to validate the use of this prototype, we conducted experiments by tourists in order to calculate indicators to assess the relevance of the recommendations provided by our system.

Medical Big Data Classification Using a Combination of Random Forest Classifier and KMeans Clustering

Abstract: An efficient classification algorithm used recently in many big data applications is the Random forest classifier algorithm. Large complex data include patient record, medicine details, and staff data etc., comprises the medical big data. Such massive data is not easy to be classified and handled in an efficient manner. Because of less accuracy and there is a chance of data deletion and also data missing using traditional methods such as Linear Classifier K-Nearest Neighbor, Random Clustering K-Nearest Neighbor. Hence we adapt the Random Forest Classification using K-means clustering algorithm to overcome the complexity and accuracy issue. In this paper, at first the medical big data is partitioned into various clusters by utilizing kmeans algorithm based upon some dimension. Then each cluster is classified by utilizing random forest classifier algorithm then it generating decision tree and it is classified based upon the specified criteria. When compared to the existing systems, the experimental results indicate that the proposed algorithm increases the data accuracy.

The Decision Model for Selection of Tourism Site Using Analytic Network Process Method

Abstract: The criteria and sub criteria-based decision model for selection of tourism site using Analytic Network Process (ANP) method was to be implemented in Yogyakarta, Indonesia. In this study, we proposed criteria and sub criteria that influenced each other and had feedback between the two so that there was a comparison of tourism site alternatives according to sub criteria and pairwise comparative assessment with scale 1-9 that was then calculated in form of matrix of pairwise comparison. The result of this study was in form of decision alternatives in form of ranking to facilitate decision makers (DMs) in finding tourism sites.

Control of an Uncertain Robot Manipulator Using an Observation-based Modified Fuzzy Sliding Mode Controller

Abstract: The main contribution of this paper is the design of a robust model reference fuzzy sliding mode observation technique to control multi-input, multi-output (MIMO) nonlinear uncertain dynamical robot manipulators. A fuzzy sliding mode controller was used in this study to control the robot manipulator in the presence of uncertainty and disturbance. To address the challenges of robustness, chattering phenomenon, and error convergence under uncertain conditions, the proposed sliding mode observer was applied to the fuzzy sliding mode controller. This theory was applied to a sixdegrees-of-freedom (DOF) PUMA robot manipulator to verify the power of the proposed method.

An Exploratory Approach to Find a Novel Metric Based Optimum Language Model for Automatic Bangla Word Prediction

Abstract: Word completion and word prediction are two important phenomena in typing that have intense effect on aiding disable people and students while using keyboard or other similar devices. Such auto completion technique also helps students significantly during learning process through constructing proper keywords during web searching. A lot of works are conducted for English language, but for Bangla, it is still very inadequate as well as the metrics used for performance computation is not rigorous yet. Bangla is one of the mostly spoken languages (3.05% of world population) and ranked as seventh among all the languages in the world. In this paper, word prediction on Bangla sentence by using stochastic, i.e. N-gram based language models are proposed for auto completing a sentence by predicting a set of words rather than a single word, which was done in previous work. A novel approach is proposed in order to find the optimum language model based on performance metric. In addition, for finding out better performance, a large Bangla corpus of different word types is used.

Multi-Swarm Whale Optimization Algorithm for Data Clustering Problems using Multiple Cooperative Strategies

Abstract: Computational Intelligence (CI) is an as of emerging area in addressing complex real world problems. The WOA has taken its root from the collective intelligent foraging behavior of humpback whales (Megaptera Novaeangliae). The standard WOA is suffers from the selection of best agent while whales searching and encircling prey. This research paper deals with the multi-swarm cooperative strategies for finding the best agents which balances the two phase‘s exploration and exploitation. The performance of invoked Multi-Swarm cooperative strategies into standard WOA i.e, MsWOA is first benchmarked on a set of 23 standard mathematical benchmark function problems which includes 7 UniModal, 6 Multi-modal and 10 fixed dimension multimodal functions. The obtained graphical and statistical results have been portrayed along with the previously established techniques. The obtained results depicts that the proposed cooperative strategies for WOA outperforms in solving optimization problems of standard benchmark functions. This paper also studies the numerical efficiency of proposed techniques on the problem of data clustering where 10 real data clustering problems have been taken from data repository https://archive.ics.uci.edu.data. Statistical results for the obtained cluster centroids, intra-cluster distances and inter-cluster distances confirms that the cooperative strategies for best whale agent selection improves the performance WOA for function optimization problems as well as data clustering problems.

Decision Support System Performance-Based Evaluation of Village Government using AHP and TOPSIS Methods: Secang Sub-district of Magelang Regency as a Case Study

Abstract: The increasing of village grants provider growth opportunities for villages every year. The Reliable performance of village governance is the main factor that determines the development of a village. In Secang, there are still many rural governance performances that are not optimal yet, therefore we need a system of performance evaluation of the village government. Decision support systems with a combination of AHP and TOPSIS models can be used to help evaluating performance village Government. AHP method is used to perform the weighting of the criteria and TOPSIS methods to make a ranking system of the performance evaluation of village government. AHP was chosen because it has many advantages of computation weighting while TOPSIS was efficient and able to measure the relative performance in a simple mathematical form. One advantage of the system that was built is the dynamic nature of the assessment criteria used for the calculation process, with menus for assessment criteria period the user can add or reduce the assessment criteria in accordance with the requirements or regulations. Output of DSS Village Government Performance Evaluation is village government performance ratings that can be used as a consideration in determining rewards and assistance to the village from the sub-district. From the test results on ranking and prototype, 86.67% of users agree that the prototype can be implemented and used to evaluate the performance of village government in the Secang sub-district.

An Improved Hybrid Distributed Collaborative Filtering Model for Recommender Engine using Apache Spark

Abstract: The present scenario there is a serious need of scalability for efficient analytics of big data. In order to achieve this, technology like MapReduce, Pig and HIVE came into action but when the question comes to scalability; Apache Spark maintains a great position far ahead. In this research paper, it has designed and developed an improved hybrid distributed collaborative model for filtering recommender engine. Execution time, scalability and robustness of the engine are the three evaluation parameters; has been considered for this present study. The present work keeps an eye on recommender system built with help of Apache Spark. Apart from this, it has been proposed and implemented the bisecting KMeans clustering algorithms. It has discussed about the comparative analysis between KMeans and Bisecting KMeans clustering algorithms on Apache Spark environment.

Solving school bus routing problem using genetic algorithm-based model

Abstract: School Bus Routing Problem is an optimization problem which falls under the class of the Vehicle Routing Problem. It involves the use of a fleet of vehicles to efficiently and optimally transport students to and from their schools. To solve this problem, optimal school bus routes are found by minimizing the number of buses, the number of routes and the total distance traversed along all routes. Manual routing of school buses have led to creation of many routes, increased number of buses and several buses navigating the same route, thereby incurring more cost. One of such methods used in solving school bus routing problems is meta-heuristic method which has proven better results in terms of optimal solution and reduced time complexity. In this study, Genetic algorithm is utilized to solve the school bus routing problem because of its simplicity and ability to generate many possible solutions. The algorithm is implemented in C# programming language and tested using secondary data obtained from Ondo State FreeSchool Bus Shuttle Scheme, Akure, Nigeria. The result shows that of all four nodes (bus stops) used in performance evaluation, Alakure to Oke-Aro junction bus stop presents as the best route which covers a total of 69 nodes with a total distance of 34.5km. This shows that there can be less number of buses in use and reduced number of routes in which the buses are assigned.

Clustering Matrix Sequences Based on the Iterative Dynamic Time Deformation Procedure

Abstract: The techniques of Dynamic Time Warping (DTW) have shown a great efficiency for clustering time series. On the other hand, it may lead to sufficiently high computational loads when it comes to processing long data sequences. For this reason, it may be appropriate to develop an iterative DTW procedure to be capable of shrinking time sequences. And later on, a clustering approach is proposed for the previously reduced data (by means of the iterative DTW). Experimental modeling tests were performed for proving its efficiency.