		FeedBack \| Contact Us \| Links \| Site Map

Welcome to Journal of Theoretical and Applied Information Technology

Submit Paper / Call for Papers

Journal receives papers in continuous flow and we will consider articles from a wide range of Information Technology disciplines encompassing the most basic research to the most innovative technologies. Please submit your papers electronically to our submission system at http://jatit.org/submit_paper.php in an MSWord, Pdf or compatible format so that they may be evaluated for publication in the upcoming issue. This journal uses a blinded review process; please remember to include all your personal identifiable information in the manuscript before submitting it for review, we will edit the necessary information at our side. Submissions to JATIT should be full research / review papers (properly indicated below main title).

Journal of Theoretical and Applied Information Technology
March 2019 | Vol. 97 No.05

Title:	DESIGN OF MODEL PREDICTIVE CONTROLLER BASED MULTI OBJECTIVE PSO AND TS MODELLING APPROACH
Author:	ALI THAMALLAH, ANIS SAKLY, FAOUZI M SAHLI
Abstract:	The performance of the predictive control scheme is related to the efficiency and cost of systems to be manipulated. However, it is intricate to express the dynamic nature of real process which is characterized by strong constraints and nonlinear terms. In this context, the present paper presents a constrained predictive control of nonlinear MIMO system. The proposed method combines the philosophies of T–S fuzzy modelling approach and a modified Dynamic Neighborhood PSO algorithm. The intelligent algorithms Particle Swam Optimization is applied to provide the control actions by solving constrained multi-objective optimization problems. At the modelling stage, T-S Fuzzy modelling technique is employed to predict the state evolutions of nonlinear process. Multivariable multi-objective predictive control of Quadruple-Tank Process based on the proposed strategy is carried out. The acquired results confirm the merit of the proposed method to deal with the control problems of nonlinear MIMO system. Furthermore, simulation tests are conducted to compare the performance of the proposed method and the multivariable predictive control with a single objective. The simulation results indicate that a multi-objective predictive strategy yields a better performance than a single objective.
Keywords:	Model Predictive Control, Multi-objective Optimization, T-S fuzzy approach, Dynamic Neighborhood PSO algorithm.
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	MULTI-LAYER FRAMEWORK FOR SECURITY AND PRIVACY BASED RISK EVALUATION ON E-GOVERNMENT
Author:	AJI SUPRIYANTO, JAZI EKO ISTIYANTO, KHABIB MUSTOFA
Abstract:	Security and privacy are an important aspect of e-Government's success in providing online services to the public. The increase of electronic service usage including e-Gov can cause various risks, safety risks, and user’s privacy risks. The lack of concern of security and privacy gives impact to some ]problems of data and information so as to make the lack of public confidence in e-Gov services. So far the concern is the security aspect, while privacy is less attention. In many cases, the privacy aspect has many violations. This study aims to develop a multi-layer security and privacy framework as a basis for the evaluation of risk-based e-Government risk awareness. The steps in this research are creating the objectives of the security and privacy framework, the identification of requirements and the relevance of requirements, constructing the inclusive security aspect, identifying of the multi-layer framework, developing the development framework, and determining the elements for the risk-based evaluation model. The contribution of this research is the compilation of a multi-layer framework model for security and privacy. The relationship between the security and privacy domains forms a complete element of security and privacy which is the development of the Salman multi-layer framework. The resulting framework can be used as a basis for conducting security based on risk evaluations involving privacy factors.
Keywords:	Framework, Requirements, Security, Privacy, Multi-layer
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	A COMPARATIVE ANALYSIS OF PHISHING WEBSITE DETECTION USING XGBOOST ALGORITHM
Author:	HAJARA MUSA, DR. A.Y GITAL, F. U. ZAMBUK, ABUBAKAR UMAR, AISHATU YAHYA UMAR, JAMILU USMAN WAZIRI
Abstract:	As most of human activities are being moved to cyberspace, phishers and other cybercriminals are making the cyberspace unsafe by causing serious risks to users and businesses as well as threatening global security and economy. Nowadays, phishers are constantly evolving new methods for luring user to reveal their sensitive information. To avoid falling victim to cybercriminals, a phishing detection algorithms is very necessary to be developed. Machine learning or data mining algorithms are used for phishing detection such as classification that categorized cyber users in to either malicious or safe users or regression that predicts the chance of being attacked by some cybercriminals in a given period of time. Many techniques have been proposed in the past for phishing detection but due to dynamic nature of some of the many phishing strategies employed by the cybercriminals, the quest for better solution is still on. In this paper, we propose a new phishing detection model based on Extreme Gradient Boosted Tree (XGBOOST) algorithm. Experimental results demonstrated that XGBOOST-based phishing detection model is promising by returning an accuracy of 97.27% which outperformed both probabilistic Neural Network (PNN) and Random forest (RF) that returned accuracies of 96.79% and 95.66% respectively.
Keywords:	Machine Learning, Feature Selection, Classification, XGBOOST, Phishing.
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	ROTATION INVARIANT FACE RECOGNITION USING JACOBI –FOURIER MOMENTS
Author:	DR ALI MOHAMMED SAHAN, ALI SAMI AZEEZ, MOHAMMED FADHIL IBRAHIM
Abstract:	The face image comprises different global and local details related to the whole and different local areas of the face image, therefore extract global features from the local areas leads to provide distinct distinguishing of the face images. In this, work a face recognition approach based on Jacobi-Fourier moments has been presented. Jacobi-Fourier moments used to provide global and distinct features about local areas in the face image. These features are used to construct histogram bins. Extensive experiments are carried out using different standard face database such as ORL, JAFEE and UMIST which are contain large variability in facial expression, pose and illumination. The experimental analysis refers that the proposed approach outperforms the traditional method under pose, facial expression, and different variations as well as it rotation invariant.
Keywords:	Rotation Invariant, Face Recognition, Jacobi-Fourier Moments, Global Features, Local Features.
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	EXTENDING THE TRANSLATION FROM PSEUDOCODE TO SOURCE CODE WITH REUSABILITY
Author:	SHEILA NURUL HUDA, ZAINUDIN ZUKHRI, TEDUH DIRGAHAYU, CHANIFAH INDAH RATNASARI
Abstract:	Pseudocode is made of a set of words in a natural language and a set of conventions to define algorithms. Pseudocode is written in a natural language that is convenience for students. In our previous work, we have developed an automatic translation from pseudocode into source code. In this paper, we extend it to handle constructs which have not been covered yet, including two types of iteration, functions, and procedures call to complete all constructs for defining algorithms. Our translation approach uses an intermediate model in XML that benefits us with the reusability of translation modules. Using reusability, we develop a new translation from pseudocode in English to source code in C++. Pseudocode and the corresponding intermediate model are Platform-independent Models (PIMs). This allows us to translate them to source code in different programming languages. The source code resulted from the translation process is Platform-Specific Model (PSM). In the translation process, it must be ensured that all models, i.e. pseudocode, the corresponding intermediate model, and the resulted source code, represent the same algorithm. Therefore, we define a conceptual metamodel for defining all the models. This paper contributes a new approach that allows reusability based on a conceptual metamodel for preserving the behavioral equivalence between all the models.
Keywords:	Pseudocode, Source Code, Conceptual Metamodel, Translation, Reusability
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	CLOUD RANKING MODEL FOR OPTIMAL SERVICE SELECTION BASED ON RANDOM FUZZY LOGIC
Author:	M. SATHEESH , M. ARAMUDHAN
Abstract:	The digital-Era generation is tuned to operate on Multi-mode resources with verity of choices of cloud services to satisfy their customers’ requirement. In the past the cloud service providers were very limited to satisfy these multi-mode customers with their inadequate available resources. Hence there is a wide increase on cloud service providers in a federated environment, due to its advantages of multiple reductions in infrastructural cost, service availability, performance and scalability. From the available cloud services, there is a great need of the hour to choose the service provider through the Cloud Broker Architecture. Along with the cloud broker architecture, the cloud service ranking model comes as an aid for an efficient selection of the cloud provider for the requested user. Hence cloud service ranking becomes a process of selecting the best and suitable provider. In this work we base our proposed method of ranking on the Fuzzy logic set. We discuss on different ranking methods and propose a suitable ranking method based on Random variable selection in ranking with the extended parameters like quality of service, cost reduction, performance and response time by the cloud service providers. In the proposed approach, Federated Cloud Architecture (FCA) shortlists the related CSPs for the user tasks automatically and chooses the optimal provider using the concept of preferential ranking mechanism. The solution is arrived by accomplishing the quality of service based SMI attributes which is the measuring parameters form the pool of cloud service provider (CSP)s.
Keywords:	Cloud service Provider (CSP), Random Fuzzy Logic, Ranking Mechanism, Quality of Service, Service Measurement Index (SMI), Service Provisioning
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	MODIFYING DES ALGORITHM BY USING DIAGONAL MATRIX BASED ON IRREDUCIBLE POLYNOMIAL
Author:	SAHAB DHEYAA MOHAMMED, ABDUL MONEM SALEH RAHMA
Abstract:	Risks of computer offenses and requirements for information confidentiality have led to increasing attention on high-security cryptosystems. Conventional encryption methods cannot provide enough security when executed on computer systems. Therefore, modern technology uses the principles of traditional encryption methods and mathematical principles applicable on computers. Data Encryption Standard (DES) must have more robust security than other cryptosystems. However, the process time necessary for cryptanalysis is less than usual. Moreover, as hardware techniques have quickly advanced, the DES may be attacked by several types of cryptanalysis using a parallel process. This study proposes changes in the operation of DES to ensure high security. Such changes include performing matrix multiplication operation instead of Exclusive OR (XOR) operation. Moreover, four keys are used for each round, two of which are derived from the main key and the remaining two are internally generated. The four keys are used in a special sequence with round numbers. The main key is generated from a random string of 64 bytes. Then, the key is expanded and distributed over 16 keys.
Keywords:	Data Encryption Standard (DES), Irreducible Polynomial, Diagonal Matrix, Polynomial
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	THE ROLE OF E-TRUST IN ACHIEVING E-LOYALTY: AN EXPLORATORY STUDY ON JORDANIAN CUSTOMERS USING SHOPPING WEBSITES
Author:	MOHAMMAD ATWAH AL-MA AITAH, ADEL ODEH AL-HASHEM
Abstract:	This study aimed to identify the state of electronic trust and its impact on electronic loyalty in the Hashemite Kingdom of Jordan. The study population consisted of Jordanian customers using shopping websites and utilized a questionnaire that was developed based on prior research and managed electronically to collect data. A convenience sample consisting of 386 individuals was used and the data was analyzed using SPSS and PLS techniques. The major findings revealed that there is a significant impact of e-trust dimensions(Credibility, Integrity and Orientation to Resolve Problems) on e-loyalty to shopping websites in the Jordanian context. The study additionally provides a list of recommendations for online retailers, such as offering precise and sufficient information on their websites and imploring the Jordanian Government to implement deterrent laws against the exploitation of customers using e-selling sites.
Keywords:	Credibility, Integrity, Orientation to resolve problems, Benevolence, E- loyalty.
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	AUTOMATED SHORELINE DETECTION DERIVED FROM VIDEO IMAGERY USING MULTI THRESHOLDING TECHNIQUES
Author:	I M.O. WIDYANTARA, I N. ARMAWAN, I M.D.P. ASANA, I B.P. ADNYANA
Abstract:	Shoreline is a zone of contact between the ocean and the land which always changes due to the movement of sediment across the coast. Dynamic changes in shoreline can cause abrasion and accretion which can damage the coastal environment. Therefore, monitoring the position of the shoreline is a very significant issue given the socio-economic value and high population density along shore areas. This paper presents a new approach in automated shorelines detection based on video images. Several sequences of image processing with the main component being image segmentation using Harmony Search Multithresholding Algorithm (HSMA) are employed. This algorithm combines the original harmony search algorithm (HSA) and the Kapur’s algorithm as an objective function to obtain the optimum threshold value in order to improve the quality of segmentation. A series of advanced image processing steps are also applied to the segmented images, mainly binarization, morphological operation, and Canny edge detection for land/sea object classification. The final product of the image processing chain is the continuous line coverage that is visualized as a shoreline. Based on the results of testing carried out on several video images, the proposed method is able to accurately detect shorelines along the borders of land and sea objects.
Keywords:	Video Monitoring, Shoreline Detection, Harmony Search Algorithm, Canny Edge Detection, Morphological Operation, Coastal Management.
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	NOVEL ALGORITHMS FOR RESOLUTION ENHANCEMENT OF IMAGES
Author:	RENU SHARMA, MADHU JAIN
Abstract:	In this paper, three new algorithms have been proposed for resolution enhancement of different gray scale images. These algorithms were based on dual tree complex wavelet transform (DTCWT). Firstly, a reference image was converted to low resolution image. In the next step, this low resolution image was decomposed using DTCWT. Due to that, high sub-band and low sub-band images had been generated. High sub-band image was further processed for sharpness enhancement. Gaussian filter and Fast non local mean (NLM) filter were further used for generating super-resolution image. Contrast limited adaptive histogram equalization (CLAHE) was also used in two of the proposed algorithms. Results were simulated on MATLAB software. Qualitative analysis proves that the visual quality of the input image is improved. Quantitative analysis was also carried out in terms of peak signal-to-noise ratio (PSNR). The simulation results show that proposed algorithms are better than existing ones.
Keywords:	Resolution Enhancement, Contrast Enhancement, DTCWT, Gaussian Filter, Fast NLM Filter.
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	USING ASSOCIATION RULE LEARNING TO DETECT DDOS ATTACK IN SPV
Author:	MARWAN A. ALBAHAR , KHALED G. MOHAMED
Abstract:	Lightweight Bitcoin clients can authenticate that a transaction is included in the blockchain, without the need of downloading the complete blockchain through Simple Payment Verification (SPV). A vast majority of lightweight clients utilizes SPV. However, with the increase in the popularity of SPV, many attacks have been launched against it such as Spoof, Sniff, Distributed Denial of Service (DDoS). As SPV requires high processing speed, flexibility, and stability, association rule learning can be effectively used for attack detection in SPV. In this paper, we implement a DDoS attack detection system using Association rule learning in SPV. Then, we used a KDD dataset to analyze the detecting accuracy and processing time in comparison with a machine learning approach. In addition, we also used NSL-KDD and GureKDD datasets to validate the effectiveness of our method. Our results show that Association rule learning algorithm is capable of detecting the DDoS attack in SPV.
Keywords:	DDos Detection, Association Rules, Cyber Security, Data Mining, SPV
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	HYBRID ALGORITHM IN IMAGE COMPRESSION BETWEEN SPATIAL DOMAIN AND FREQUENCY DOMAIN BASED ON DISCRETE COSINE TRANSFORM
Author:	ZYAD THALJI
Abstract:	Image compression comprises a method for decreasing image size. In the image compression process, the redundancy in a matrix code is removed. Compression allows data to be efficiently transmitted or stored. This paper demonstrates the integration of the spatial domain and frequency domain in resizing image dimensions according to the pixel location. For image coding, the Discrete Cosine Transform (DCT) was employed in the frequency domain. Then, for coding and compressing the image, Huffman coding was used. During the first level, quantization was employed. Before using the Huffman decoding, Dequantization was used for image decoding. This prepares the matrix for the use of the Inverse Discrete Cosine Transform (IDCT), in the frequency domain for image decoding. The new algorithm is efficient, producing a high rate in image compression.
Keywords:	Image Compression, Discrete Cosine Transform, quantization, Spatial Domain, Frequency Domain
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	EVOLUTIONARY MODEL FOR THE ITERATED N-PLAYERS PRISONERS’ DILEMMA BASED ON PARTICLE SWARM OPTIMIZATION
Author:	SALLY ALMANASRA
Abstract:	Evolving the cooperative behavior in Iterated N-Players Prisoners’ Dilemma (INPPD) is studied over several evolutionary models. These models presented solutions for evolving cooperative behavior among INPPD players. Studying existing models revealed that when the number of the players’ increases, the models lose their capabilities in maintaining stable levels of cooperation between the players. In this paper, we present an evolutionary model for enhancing the cooperation levels in large population of INPPD players. The model focuses on optimizing the communication topology of INPPD, as well as building a knowledge base to support players’ future decisions based on the evolved knowledge gained from historical actions taken by different players. The presented communication topology along with the knowledge base present considerable support for the evolutionary Particle Swarm Optimization (PSO) algorithm to evolve the players’ strategies. The results showed that the model could increase the cooperative rate among INPPD player and allow players to achieve higher payoffs against benchmark strategies
Keywords:	Prisoners’ Dilemma, Game Theory, Communication Topology, Particle Swarm Optimization, Knowledge base
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	IMPROVE IMAGE REGISTRATION JEFFREY’S DIVERGENCE METHOD FOR INSUFFICIENT OVERLAP AREA USING KMEANS++ IN REMOTE SENSED IMAGES
Author:	MOHAMMAD AWWAD ALNAGDAWI, SITI MARIYAM HJ.SHAMSUDDIN , SITI ZAITON MOHD HASHIM, ALA ABURUMMAN
Abstract:	In remote sensing, lacking sufficient overlap area is a common problem for image registration. To address this issue, Jeffrey’s divergence intensity-based registration technique was developed. This technique is not robust enough when dealing with multimodal images because it influences by the amount of variance in the data, so it may fail to find the optimal registration. Image segmentation can help to reduce the difference between the multimodal images while keep the salient features. kmeans++ was adopted for image segmentation because of it simple and efficient. This segmentation help Jeffrey’s divergence to be more robust with local intensity variation and get the optimal registration even with smaller overlap area. Comprehensive results were conduct to shows the impact of the proposed method to get a better result to compare with the state-of-the-art methods, Jeffrey’s divergence (JD) and mutual information (MI).
Keywords:	K-Means++; Jeffrey’s Divergence; Multimodal Image Registration; Mutual Information; Remote Sensing Image;
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	CONSUMERS CULTURAL ATTITUDE TOWARD MOBILE ADVERTISING: AN EMPIRICAL INVESTIGATION AMONG DIFFERENT NATIONAL CULTURES
Author:	MAIDUL ISLAM, MINCHEOL KANG
Abstract:	The aim of this study is to examine the relationships between the characteristics of advertisement and consumer attitude toward mobile advertisement. How national culture moderates those relationships. A survey was conducted with consumers from three different countries (Korea, India, and China). Results show that consumers are willing to accept mobile ads if they are informative, credible, and interactive. Results also indicate that the Power Distance Index (PDI) of national culture has positive moderating effects on the relationships of informativeness and credibility, with attitude toward mobile advertising and the Uncertainty Avoidance Index (UAI) of national culture having negative moderating effects on the interactivity of mobile ads.
Keywords:	National culture, Hofstede, Mobile Advertising, Consumer Attitude, Cultural Dimension
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	VEHICULAR QUEUE LENGTH MEASUREMENT BASED ON EDGE DETECTION AND VEHICLE FEATURE EXTRACTION
Author:	WAHBAN AL OKAISHI, ABDELMOGHIT ZAARANE, IBTISSAM SLIMANI, ISSAM ATOUF, MOHAMED BENRABH
Abstract:	The queue length is an important parameter used by several companies of traffic management for quantitative analysis of traffic scene. In this paper, we propose a system to measure the queue length. The proposed system contains two principal operations for measuring the queue length the first operation is the motion detection, and the second one is the vehicle detection. The traffic scene is divided into a number of blocks, which have variable sizes based on the camera parameters and the distance between the camera and the position of the block. To minimize the execution time to accord the real-time application, we continuously apply the detection motion operation on the first block of the scene and the queue tail. The vehicle detection algorithm is based on the edge detection and the vehicle features extraction to improve the detection of vehicle and minimize the error of detection of the other things’ edges (damaged of road, the mark of the road, the shadow of trees or building). The algorithm is applied to videos obtained by stationary camera. The obtained results demonstrate the robustness and accuracy of the proposed system.
Keywords:	Vehicular Queue Length, Image Processing, Edge Detection, Vehicle Feature Extraction
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	AUTOMATED COMPLAINTS CLASSIFICATION USING MODIFIED NAZIEF-ADRIANI STEMMING ALGORITHM AND NAIVE BAYES CLASSIFIER
Author:	VANNIA FERDINA, MARCEL BONAR KRISTANDA, SENG HANSUN
Abstract:	Complaints provided by customers in the use of products or services is a feedback of the quality of products or services used by customers. In Universitas Multimedia Nusantara (UMN), students can deliver their complaints through an organization, i.e. Dewan Keluarga Besar Mahasiswa (DKBM) UMN. All students’ complaints are manually classified into predefined categories by DKBM so that it can be delivered to related division. It costs a lot of time and human resources of DKBM UMN, and also caused misclassification of incoming complaints. In e-complaint system, a method that can be used to support efficient complaint processing is the use of automatic classification system because it can save both time and human resources. Naive Bayes Classifier (NBC) algorithm is one the algorithm that can be used to classify text automatically and for the preprocessing stage, modified Nazief-Adriani stemming algorithm is used. Based on the study conducted, it can be concluded that Naive Bayes Classifier algorithm with modified Nazief-Adriani stemming algorithm is able to do the classification well. This is indicated from the precision value of 91.86%, the recall value of 84.48%, and the f-1 score value of 86.29% for the ratio of training data and test data 90:10, and an average accuracy of 86%.
Keywords:	e-Complaint, Naive Bayes, Classifier Algorithm, Text Classification, Text Mining
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	TEXT IN IMAGE STEGANOGRAPHY BASED ON A DYNAMIC NON-SEQUENTIAL LEAST SIGNIFICANT BIT TECHNIQUE IN GRAYSCALE AND RGB IMAGES
Author:	HAYFAA ABDULZAHRA ATEE, ABIDULKARIM K.I.YASARI, ROBIAH AHMAD, NORLIZA MOHD NOOR
Abstract:	Information hiding is one of the considerable objects in secret communication. Steganography is a vital research area in recent years relating several applications. Image steganography is the mode of embedding information (e.g. text) in an image, such an impossible to be seen by human eyes or so-called visual system (HVS). This study presents a steganography scheme based on a dynamic non-sequential Least Significant Bit (LSB) procedure for concealing text information into two kinds of images; include RGB images and Grayscale images, where the efficiency of using each of these kinds is studied. the spatial domain is used to perform the LSB procedure, where the bits of the secret message are inserted into the host or cover image using LSB process to outcome a stego-image. The results show that the secret message can be hidden securely and undetectable against HVS, and using RGB image shows better performance. Comparing with some other methods, the proposed scheme is superior in terms of MSE and PSNR.
Keywords:	LSB, Image steganography, PSNR, MSE, HVS.
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	PROOF OF WORK: ENERGY INEFFICIENCY AND PROFITABILITY
Author:	ANAK AGUNG GDE AGUNG, RIXARD G. DILLAK, DEVIE R. SUCHENDRA, ROBBI H
Abstract:	Decentralized and immutable characteristic of blockchain has a possibility to change how the data is stored. Cryptocurrency is one example of successful blockchain technology implementation. The first cryptocurrency, bitcoin, was launched in 2009 and shortly afterwards followed by other cryptocurrencies, which are called alternative currency (altcoin). The blockchain system depends on a consensus mechanism to run. Most of cryptocurrency adopt the Proof of Work (PoW) consensus mechanism, which requires to run a computer program to solve a computational puzzle to verify the transactions and add the record into the blockchain, which called mining. Bitcoin uses SHA258 algorithm for its PoW. As an incentive, miners are then given some money on the currency. However, mining requires a lot of energy, alternatively, altcoins adopt different algorithm to run the system. This study aims to compare the energy used by various algorithms, which mined by four widely available, general purposes Graphic Processing Unit (GPU), and determine the profitability for each currency, given the mining share acquired for 24 hours. This is important because even the blockchain is not intended primarily for cryptocurrency, PoW-based blockchain system depends heavily on the mining process. Should the miners decided it is no longer profitable, they will easily switch to mine another, and without miners, the blockchain system will stop. The experiment shows that from 32 sets of experiment, only 15 sets (46.88%) are profitable. The result shows that among eight algorithms, Equihash, Ethash, and Cryptonight7 coins are the best performers, while Blake2b, Blake256, and Lyra2REv2 coins are the worst performers. Most the coins tested consume below than 1 TWh of annual energy consumption, except SiaCoin and Ethereum, and Decred.
Keywords:	Cryptocurrency, Altcoin, Proof-Of-Work, Energy, Profitability
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	DESIGN DEEP LEARNING NEURAL NETWORK FOR STRUCTURAL HEALTH MONITORING
Author:	REZA RAHUTOMO, FERGYANTO E. GUNAWA
Abstract:	Bridge structural failure happens as the lack of monitoring. The existence of bridge structural health monitoring system is necessary for bridge maintenance due to its ability to process data and provide the information of structural health level. This research is performed to design a deep neural network model for classifying structural integrity with high accuracy. The model requires input data in the form of F-statistic, which is derived from structural vibration data. In the current approach, the vibration data are obtained from numerical analysis by means of the finite element methods. As much as 17.493 vibration cases are generated for five levels of structural integrity, namely, healthy conditions and conditions of 1%, 5%, 10%, 20% damage level. The neural network model consists of one input layer of 20 neurons, six hidden layers with 12 neurons per layer, and one output layer of 5 neurons. The model is trained by using Adam optimizer. The results show that the model is able to accurately classify the structural damage at 83.3% accuracy, and the majority of the false predictions occur in differentiating the healthy structural condition from those of 1% damage.
Keywords:	Artificial Neural Network, Deep Learning, Structural Health Monitoring, Vibration-based, Classification Accuracy
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text

Title:	AN EMPIRICAL COMPARISON OF SOME MODIFIED NEAREST NEIGHBOR RULE FOR CREDIT SCORING ANALYSIS: CASE STUDY IN INDONESIA
Author:	MOCH. ABDUL MUKID, TATIK WIDIHARIH, MUSTAFID
Abstract:	This paper aims to examine some nonparametric classification methods based on the nearest neighbors rule including k nearest neighbors (KNN), distance weighted k nearest neighbors (DWKNN), local mean k nearest neighbors (LMKNN), and pseudo nearest neighbors (PNN). In order to know the performance of each method, we apply it in case of credit scoring in Indonesia, especially related to a micro credit. For each method studied, we use the same parameter, i.e. the Euclidean distance. After evaluating some odd value k, it is known that each method achieves the optimum classification performance at different k values. KNN achieved the best performance value at k = 11 with total accuracy of 84.91%, while DWKNN achieved best performance at k = 15 which only reached 77.36%. LMKNN works well on k = 9 with an accuracy value of 84.91% and PNN which is a combination between DWKNN and LMKNN only has an accuracy classification of 83.02%. In the case of micro credit in Indonesia with samples from a government bank in Wonogiri district, LMKNN is able to perform better than other methods. With k = 9, the classification performance of LMKNN is the same with the KNN that is obtained at k = 11. Therefore by using LMKNN will reduce the time in determining the label class of a prospective borrower.
Keywords:	Nonparametric Classification, K Nearest Neighbors, Distance Weighted K Nearest Neighbors, Local Mean K Nearest Neighbors, And Pseudo Nearest Neighbors
Source:	Journal of Theoretical and Applied Information Technology 15^th March 2019 -- Vol. 97. No. 05 -- 2019
Full Text