International Journal of Computer (IJC) 2021-05-11T17:45:46+00:00 Dr. Mohamad Nasar Open Journal Systems <p>The <a title="International Journal of Computer (IJC) home page" href="/index.php/InternationalJournalOfComputer/index" target="_blank" rel="noopener"><strong>International Journal of Computer (IJC)</strong></a> is an open access International Journal for scientists and researchers to publish their scientific papers in Computer Science related fields. <a title="International Journal of Computer (IJC)" href="/index.php/InternationalJournalOfComputer/index" target="_blank" rel="noopener">IJC</a> plays its role as a refereed international journal to publish research results conducted by researchers.</p> <p>This journal accepts scientific papers for publication after passing the journal's double&nbsp;peer review process.&nbsp; For detailed information about the journal kindly check <a title="About the Journal" href="/index.php/InternationalJournalOfComputer/about">About the Journal</a>&nbsp;page.&nbsp;</p> <p>All <a title="International Journal of Computer (IJC)" href="/index.php/InternationalJournalOfComputer/index" target="_blank" rel="noopener">IJC</a> published papers in Computer Science will be available for scientific readers for free; no fees are required to download published papers in this international journal.</p> <p>&nbsp;</p> Presence Detection with Bluetooth Low Energy: A Review and Experiment 2021-02-11T10:27:34+00:00 Michael Hosein Kevin Jaglal <p>Bluetooth is one of the most ubiquitous technologies in smart phone today and its prominence in other devices is rising rapidly. It has become the De Facto technology used when there is need for device-to-device communication. However, the evolving standard has much more to offer. Bluetooth can power many applications due to capabilities. A key metric of Bluetooth is the Received Signal Strength Indicator (RSSI) and depending on the readings one can infer locality. This study evaluates existing research that attempts localization implemented using the Bluetooth protocol and the metrics that power those applications. A proof-of-concept software is developed to further investigate the feasibility of presence detection using Bluetooth Low Energy without connection to a device.&nbsp;</p> 2021-01-29T19:30:37+00:00 Copyright (c) 2021 International Journal of Computer (IJC) Elaboration of Processing Chains for Spatio-temporal Analysis of Rainfall Data Application: Alaotra Mangoro Region, Madagascar 2021-02-23T07:31:03+00:00 Aimé Richard Hajalalaina Arisetra Razafinimaro Niry Arinavalona Rakotovao Adolphe Ratiarison <p>This present study was carried out within the framework of the automation of the processing chains of spatio-temporal data processing of climatological data in Madagascar. Our objective is to develop and automate the processing chains of rainfall data from ERA-Interim re-analyses of the European Meteorological Centre ECMWF to observe and analyze the evolution of precipitation over time and space. The chains are elaborated in a generic way by introducing statistical methods such as spatial interpolation, calculation of temporal and spatial averages, maximum entropy method, and principal component analysis. Then, the elaborated chains are implemented with MATLAB. Thus, the test of these chains has been performed in the Alaotra Mangoro region. The results obtained allowed us to observe, analyze and interpret the evolution of precipitation in this region of Madagascar during 35 years (1979 to 2013).</p> 2021-02-23T07:30:37+00:00 Copyright (c) 2021 International Journal of Computer (IJC) Design of a MPPT System Based on Modified Grey Wolf Optimization Algorithm in Photovoltaic System under Partially Shaded Condition 2021-03-01T06:46:28+00:00 Muhammad Ilyas Hatem Khalifa Emhmed Ghazal <p>Conventional Maximum Potential Monitoring strategies such as perturbation and observation, incremental conduct, and climbing can effectively monitor the maximum power point in uniform shading, whereas failing in a partially shaded condition. Nevertheless, it is difficult to achieve optimal and reliable power by using photovoltaics. So, to solve this issue, this article proposes to monitor the photovoltaic system's global optimum powerpoint for partial shading with a Modified Gray Wolf Optimizer (MGWO) based maximum power point tracking algorithm. Under partial shadows, a mathematical model of the PV system is built with a single diode, EGWO is used to monitor global maximum power points.&nbsp; A photovoltaic system includes deciding which converter is used to increase photovoltaic power generation. The MPPT architecture uses a modified gray wolf optimization algorithm to quickly track the output power and reduce photovoltaic oscillations. The efficiency of the maximum power tracker is better than the GWO algorithm of up to 0,4 s with the modified gray wolf optimization algorithm. Converters are used to resolve the power losses often occurring in PV systems with a soft-buck converter process.&nbsp; The output of the power generator is greater than the soft-switching buck converter. The simulation and experimental results obtained suggest that both the P &amp; O and IPSO MPPTs are superior to the proposed MPPT algorithm, the proposed algorithm increases the traceability efficiency. The suggested algorithm has the fastest follow-up speed since the α value decreases during the iteration exponentially.</p> 2021-03-01T06:46:28+00:00 Copyright (c) 2021 International Journal of Computer (IJC) Marigold Blooming Maturity Levels Classification Using Machine Learning Algorithms 2021-03-07T08:33:27+00:00 S M Abdullah Al Shuaeb Shamsul Alam Mohammod Hazrat Ali Md. Kamruzzaman <p>Image processing is swiftly progressive in the area of computer science and engineering. Image classification is a fascinating task in image processing. In this study, we have classified the marigold blooming maturity levels like a marigold bud, partial blooming marigold, and fully blooming marigold. To classify the marigold blooming maturity levels are a tough and time-consuming task for human beings. Hence, an automatic marigold maturity levels classification tool is very adjuvant even for experience humans to classify the huge number of marigolds. For the sake of that, we have deliberated a novel system to classify automatically marigold blooming maturity levels image data by using machine learning algorithms. There are three types of machine learning models namely Artificial Neural Network(ANN), Convolutional Neural Network(CNN), and Support Vector Machine(SVM) that are used to automatically classify marigold maturity levels. Hence, we have preprocessed the image at first. Then we extract the various features from the marigold images. After that, these features have fed into Machine Learning(ML) models and classify these images into the category. From the experiment, we observed that the Convolutional Neural Network (CNN) model provides a high accuracy compared to other Artificial Neural Network(ANN) and Support Vector Machine(SVM) algorithms. The Convolutional Neural Network(CNN) models performed the best among all two classifiers with an overall accuracy of 93.9%. Our proposed system is efficiently classifying marigold maturity levels.</p> 2021-03-06T00:00:00+00:00 Copyright (c) 2021 International Journal of Computer (IJC) Design of a Digital Game-Based Learning Environment for Solving Quadratic Equations Using Completing-the-Square-Method 2021-04-10T05:46:13+00:00 Oluwatoyin Catherine Agbonifo Akintoba Emmanuel Akinwonmi Mosope Samuel Williamson <p>Various preconceptions about the effectiveness of applying digital-games approach in tandem with traditional teaching methods subsist in spite of learners increasing usage of digital devices and digital games. This trend obviously underplays the existing technological advancements made in respect to digital devices and computer game programming. This research paper applied the digital game approach to the teaching of mathematics with a view to boosting learner’s interest while mitigating boredom, difficulty and apprehension towards solving problems. It employed story-telling technique and role-play (both fun elements) to mathematics learning while still preserving the traditional stepwise approach to problem-solving in mathematics. A digital game-based environment was developed based on the battleship game. This environment was used to learn how to solve the quadratic equation using completing the square method. Performance evaluation was carried out to determine if the system aligns with the underlined objectives. The findings showed that using the digital game-based learning system helps in reducing learners’ apprehension in solving the quadratic problem and improved their cognitive skills in solving quadratic equations.</p> 2021-04-10T05:46:13+00:00 Copyright (c) 2021 International Journal of Computer (IJC) Metadata Extraction from References of Different Styles 2021-05-04T20:15:55+00:00 Olugbenga A Madamidola Olatunde, Ibikunle Olawale T Adeboje Promise I Ayansola <p>Metadata extraction is the process of describing extrinsic and intrinsic qualities of the resource such as document, image, video, including getting data from references. References form an essential part of electronic scholarly publications. A reference is the way of giving acknowledgment to individuals for their creative and intellectual works that one utilized in his or her research work. It can also be used to locate particular sources and combat plagiarism. A reference style dictates the information necessary for a reference and how the information is ordered. Accurate and automatic reference metadata generation provides scalability, interoperability and usability for digital libraries of both public and private institution and their collections. Accurate reference metadata extraction becomes an intriguing task to researchers who want to collect data of scientific publications; therefore, this research work proposes a metadata extraction from references of different styles with the use of regular expression. This work accurately extract metadata such as author, title of article, volume, year of publication and institution from references of different styles limiting it to six referencing style<strong><em>.</em></strong></p> 2021-05-04T20:15:54+00:00 Copyright (c) 2021 International Journal of Computer (IJC) Peer-to-peer Approach for Distributed Privacy-preserving Deep Learning 2021-05-11T17:45:46+00:00 Mustapha Abdulkadir Sani Abdulmalik A. Lawan Salisu Mamman. Abdulrahman <p>The revolutionary advances in machine learning and Artificial Intelligence have enables people to rethink how we integrate information, analyze data, and use the resulting insights to improve decision making. Deep learning is the most effective, supervised, time and cost efficient machine learning approach which is becoming popular in building today’s applications such as self-driving cars, medical diagnosis systems, automatic speech recognition, machine translation, text-to-speech conversion and many others. On the other hand the success of deep learning among others depends on large volume of data available for training the model. Depending on the domain of application, the data needed for training the model may contain sensitive and private information whose privacy needs to be preserved. One of the challenges that need to be address in deep learning is how to ensure that the privacy of training data is preserved without sacrificing the accuracy of the model. In this work, we propose, design and implement a decentralized deep learning system using peer-to-peer architecture that enables multiple data owners to jointly train deep learning models without disclosing their training data to one another and at the same time benefit from each other’s dataset through exchanging model parameters during the training. We implemented our approach using two popular deep learning frameworks namely Keras and TensorFlow. We evaluated our approach on two popular datasets in deep learning community namely MNIST and Fashion-MNIST datasets. Using our approach, we were able to train models whose accuracy is relatively close to models trained under privacy-violating setting, while at the same time preserving the privacy of the training data.</p> 2021-05-11T17:45:46+00:00 Copyright (c) 2021 International Journal of Computer (IJC)