Open Access Published by De Gruyter Open Access

Volume 12 Issue 1

January 2022

Issue of Open Computer Science

Contents
Journal Overview

Regular Articles

Open Access February 21, 2022

An ROI-based robust video steganography technique using SVD in wavelet domain

Urmila Pilania, Rohit Tanwar, Prinima Gupta

Page range: 1-16

More Download PDF

Abstract

Steganography is a technique that embeds secret information in a suitable cover file such as text, image, audio, and video in such a manner that secret information remains invisible to the outside world. The study of the literature relevant to video steganography reveals that a tradeoff exists in attaining the acceptable values of various evaluation parameters such as a higher capacity usually results in lesser robustness or imperceptibility. In this article, we propose a technique that achieves high capacity along with required robustness. The embedding capacity is increased using singular value decomposition compression. To achieve the desired robustness, we constrain the embedding of the secret message in the region of interest in the cover video file. In this manner, we also succeed in maintaining the required imperceptibility. We prefer Haar-based lifting scheme in the wavelet domain for embedding the information because of its intrinsic benefits. We have implemented our suggested technique using MATLAB. The analysis of results on the prespecified parameters of the steganography justifies the effectiveness of the proposed technique.

Open Access February 23, 2022

SIKM – a smart cryptographic key management framework

Saman Shojae Chaeikar, Ali Ahmadi, Sasan Karamizadeh, Nakisa Shoja Chaeikar

Page range: 17-26

More Download PDF

Abstract

For a secure data transmission in symmetric cryptography, data are encrypted and decrypted using an identical key. The process of creating, distributing, storing, deploying, and finally revoking the symmetric keys is called key management. Many key management schemes are devised that each one is suitable for a specific range of applications. However, these schemes have some common drawbacks like the hardness of key generation and distribution, key storage, attacks, and traffic load. In this article, a key management framework is proposed, which is attack resistant and transforms the current customary key management workflow to enhance security and reduce weaknesses. The main features of the proposed framework are eliminating key storage, smart attack resistant feature, reducing multiple-times key distribution to just one-time interpreter distribution, and having short key intervals – minutely, hourly, and daily. Moreover, the key revocation process happens automatically and with no revocation call.

Open Access February 23, 2022

Predicting and monitoring COVID-19 epidemic trends in India using sequence-to-sequence model and an adaptive SEIR model

Koyel Datta Gupta, Rinky Dwivedi, Deepak Kumar Sharma

Page range: 27-36

More Download PDF

Abstract

In the year 2019, during the month of December, the first case of SARS-CoV-2 was reported in China. As per reports, the virus started spreading from a wet market in the Wuhan City. The person infected with the virus is diagnosed with cough and fever, and in some rare occasions, the person suffers from breathing inabilities. The highly contagious nature of this corona virus disease (COVID-19) caused the rapid outbreak of the disease around the world. India contracted the disease from China and reported its first case on January 30, 2020, in Kerala. Despite several counter measures taken by Government, India like other countries could not restrict the outbreak of the epidemic. However, it is believed that the strict policies adopted by the Indian Government have slowed the rate of the epidemic to a certain extent. This article proposes an adaptive SEIR disease model and a sequence-to-sequence (Seq2Seq) learning model to predict the future trend of COVID-19 outbreak in India and analyze the performance of these models. Optimization of hyper parameters using RMSProp is done to obtain an efficient model with lower convergence time. This article focuses on evaluating the performance of deep learning networks and epidemiological models in predicting a pandemic outbreak.

Open Access February 24, 2022

3D chaotic map-cosine transformation based approach to video encryption and decryption

Mohit Dua, Drishti Makhija, Pilla Yamini Lakshmi Manasa, Prashant Mishra

Page range: 37-56

More Download PDF

Abstract

Data security is vital for multimedia communication. A number of cryptographic algorithms have been developed for the secure transmission of text and image data. Very few contributions have been made in the area of video encryption because of the large input data size and time constraints. However, due to the massive increase in digital media transfer within networks, the security of video data has become one of the most important features of network reliability. Block encryption techniques and 1D-chaotic maps have been previously used for the process of video encryption. Although the results obtained by using 1D-chaotic maps were quite satisfactory, the approach had many limitations as these maps have less dynamic behavior. To overcome these drawbacks, this article proposes an Intertwining Logistic Map (ILM)-Cosine transformation-based video encryption technique. The first step involved segmenting the input video into multiple frames based on the frames per second (FPS) value and the length of the video. Next, each frame was selected, and the correlation among the pixels was reduced by a process called permutation/scrambling. In addition, each frame was rotated by 90° in the anticlockwise direction to induce more randomness into the encryption process. Furthermore, by using an approach called the random order substitution technique, changes were made in each of the images, row-wise and column-wise. Finally, all the encrypted frames were jumbled according to a frame selection key and were joined to generate an encrypted video, which was the output delivered to the user. The efficiency of this method was tested based on the state of various parameters like Entropy, Unified Average Change in Intensity (UACI), and correlation coefficient (CC). The presented approach also decrypts the encrypted video, and the decryption quality was checked using parameters such as mean square error (MSE) and peak signal-to-noise ratio (PSNR).

Open Access February 24, 2022

Security and privacy issues in federated healthcare – An overview

Jansi Rani Amalraj, Robert Lourdusamy

Page range: 57-65

More Download PDF

Abstract

Securing medical records is a significant task in Healthcare communication. The major setback during the transfer of medical data in the electronic medium is the inherent difficulty in preserving data confidentiality and patients’ privacy. The innovation in technology and improvisation in the medical field has given numerous advancements in transferring the medical data with foolproof security. In today’s healthcare industry, federated network operation is gaining significance to deal with distributed network resources due to the efficient handling of privacy issues. The design of a federated security system for healthcare services is one of the intense research topics. This article highlights the importance of federated learning in healthcare. Also, the article discusses the privacy and security issues in communicating the e-health data.

Open Access March 3, 2022

Designing of fault-tolerant computer system structures using residue number systems

Victor Krasnobayev, Alexandr Kuznetsov, Anastasiia Kiian

Page range: 66-74

More Download PDF

Abstract

This article discusses computing systems that operate in residue number systems (RNSs). The main direction of improving computer systems (CSs) is increasing the speed of implementation of arithmetic operations and the reliability of their functioning. Encoding data in RNS solves the problem of optimal redundancy, i.e., the creation of such computing systems provides maximum reliability with restrictions on weight and size characteristics. This article proposes new structures of fault-tolerant CSs operating in RNS in the case of the application with an active fault-tolerant method. The use of the active fault-tolerant method (dynamic redundancy) in the RNSs provides higher reliability. In addition, with an increase in the digits of CSs, the efficiency of using the proposed structures increases.

Open Access March 8, 2022

A method for detecting objects in dense scenes

Chuanyun Xu, Yu Zheng, Yang Zhang, Gang Li, Ying Wang

Page range: 75-82

More Download PDF

Abstract

Recent object detectors have achieved excellent performance in accuracy and speed. Even with such impressive results, the most advanced detectors are challenging in dense scenes. In this article, we analyze and find the reasons for the decrease in detection accuracy in dense scenes. We started our work in terms of region proposal and location loss. We found that low-quality proposal regions during the training process are the main factors affecting detection accuracy. To prove our research, we established and trained a dense detection model based on Cascade R-CNN. The model achieves an accuracy of mAP 0.413 on the SKU-110K sub-dataset. Our results show that improving the quality of recommended regions can effectively improve the detection accuracy in dense scenes.

Open Access March 10, 2022

An effective integrated machine learning approach for detecting diabetic retinopathy

Penikalapati Pragathi, Agastyaraju Nagaraja Rao

Page range: 83-91

More Download PDF

Abstract

Millions of people across the world are suffering from diabetic retinopathy. This disease majorly affects the retina of the eye, and if not identified priorly causes permanent blindness. Hence, detecting diabetic retinopathy at an early stage is very important to safeguard people from blindness. Several machine learning (ML) algorithms are implemented on the dataset of diabetic retinopathy available in the UCI ML repository to detect the symptoms of diabetic retinopathy. But, most of those algorithms are implemented individually. Hence, this article proposes an effective integrated ML approach that uses the support vector machine (SVM), principal component analysis (PCA), and moth-flame optimization techniques. Initially, the ML algorithms decision tree (DT), SVM, random forest (RF), and Naïve Bayes (NB) are applied to the diabetic retinopathy dataset. Among these, the SVM algorithm is outperformed with an average of 76.96% performance. Later, all the aforementioned ML algorithms are implemented by integrating the PCA technique to reduce the dimensions of the dataset. After integrating PCA, it is noticed that the performance of the algorithms NB, RF, and SVM is reduced dramatically; on the contrary, the performance of DT is increased. To improve the performance of ML algorithms, the moth-flame optimization technique is integrated with SVM and PCA. This proposed approach is outperformed with an average of 85.61% performance among all the other considered ML algorithms, and the classification of class labels is achieved correctly.

Open Access March 15, 2022

Greatest-common-divisor dependency of juggling sequence rotation efficient performance

Joseph A. Erho, Juliana I. Consul, Bunakiye R. Japheth

Page range: 92-102

More Download PDF

Abstract

In previous experimental study with three-way-reversal and juggling sequence rotation algorithms, using 20,000,000 elements for type LONG in Java, the average execution times have been shown to be 49.66761ms and 246.4394ms, respectively. These results have revealed appreciable low performance in the juggling algorithm despite its proven optimality. However, the juggling algorithm has also exhibited efficiency with some offset ranges. Due to this pattern of the juggling algorithm, the current study is focused on investigating source of the inefficiency on the average performance. Samples were extracted from the previous experimental data, presented differently and analyzed both graphically and in tabular form. Greatest common divisor values from the data that equal offsets were used. As emanating from the previous study, the Java language used for the rotation was to simulate ordering of tasks for safety and efficiency in the context of real-time task scheduling. Outcome of the investigation shows that juggling rotation performance competes favorably with three-way-reversal rotation (and even better in few cases) for certain offsets, but poorly with the rests. This study identifies the poorest performances around offsets in the neighborhood of square root of the sequence size. From the outcome, the study therefore strongly advises application developers (especially for real-time systems) to be mindful of where and how to in using juggling rotation.

Open Access March 16, 2022

Construction of a gas condensate field development model

Alexander K. Skiba

Page range: 103-111

More Download PDF

Abstract

This article has developed and verified a mathematical aggregated approximate model of developing a gas condensate field using a cyclic process. The essence of the cyclic process is to pump the drained gas into the productive formation to reduce the pressure drop into the deposit. This process allows for increased condensate recovery in the future. The model discussed in this article is a continuous dynamic system with control parameters. It is a modification of the dynamic aggregated model of a purely gas field, designed for planning for a sufficiently long period with limited information about the state of the reservoir (the initial flow rate of wells, the initial recoverable gas reserves, the initial reservoir pressure, the dependence of potential condensate content per unit volume of fatty gas on the reservoir pressure). A non-standard approach underlies the model construction. Logical simplifications and a priori assumptions about the processes occurring in the field during its development are at its core. The instruments in the model are the increase in production and injection wells and the proportion of injection wells involved in the production. The purpose of the article is to calculate various variants of the dynamics of the fundamental indicators of the development of a gas condensate field for a sufficiently long-term period at the stage of preliminary design.

Open Access March 19, 2022

A novel similarity measure of link prediction in bipartite social networks based on neighborhood structure

Fariba Sarhangnia, Shima Mahjoobi, Samaneh Jamshidi

Page range: 112-122

More Download PDF

Abstract

Link prediction is one of the methods of social network analysis. Bipartite networks are a type of complex network that can be used to model many natural events. In this study, a novel similarity measure for link prediction in bipartite networks is presented. Due to the fact that classical social network link prediction methods are less efficient and effective for use in bipartite network, it is necessary to use bipartite network-specific methods to solve this problem. The purpose of this study is to provide a centralized and comprehensive method based on the neighborhood structure that performs better than the existing classical methods. The proposed method consists of a combination of criteria based on the neighborhood structure. Here, the classical criteria for link prediction by modifying the bipartite network are defined. These modified criteria constitute the main component of the proposed similarity measure. In addition to low simplicity and complexity, this method has high efficiency. The simulation results show that the proposed method with a superiority of 0.5% over MetaPath, 1.32% over FriendLink, and 1.8% over Katz in the f -measure criterion shows the best performance.

Open Access March 21, 2022

Rough set-based entropy measure with weighted density outlier detection method

Tamilarasu Sangeetha, Amalanathan Geetha Mary

Page range: 123-133

More Download PDF

Abstract

The rough set theory is a powerful numerical model used to handle the impreciseness and ambiguity of data. Many existing multigranulation rough set models were derived from the multigranulation decision-theoretic rough set framework. The multigranulation rough set theory is very desirable in many practical applications such as high-dimensional knowledge discovery, distributional information systems, and multisource data processing. So far research works were carried out only for multigranulation rough sets in extraction, selection of features, reduction of data, decision rules, and pattern extraction. The proposed approach mainly focuses on anomaly detection in qualitative data with multiple granules. The approximations of the dataset will be derived through multiequivalence relation, and then, the rough set-based entropy measure with weighted density method is applied on every object and attribute. For detecting outliers, threshold value fixation is performed based on the estimated weight. The performance of the algorithm is evaluated and compared with existing outlier detection algorithms. Datasets such as breast cancer, chess, and car evaluation have been taken from the UCI repository to prove its efficiency and performance.

Open Access March 24, 2022

Word2Vec: Optimal hyperparameters and their impact on natural language processing downstream tasks

Tosin Adewumi, Foteini Liwicki, Marcus Liwicki

Page range: 134-141

More Download PDF

Abstract

Word2Vec is a prominent model for natural language processing tasks. Similar inspiration is found in distributed embeddings (word-vectors) in recent state-of-the-art deep neural networks. However, wrong combination of hyperparameters can produce embeddings with poor quality. The objective of this work is to empirically show that Word2Vec optimal combination of hyper-parameters exists and evaluate various combinations. We compare them with the publicly released, original Word2Vec embedding. Both intrinsic and extrinsic (downstream) evaluations are carried out, including named entity recognition and sentiment analysis. Our main contributions include showing that the best model is usually task-specific, high analogy scores do not necessarily correlate positively with F 1 scores, and performance is not dependent on data size alone. If ethical considerations to save time, energy, and the environment are made, then relatively smaller corpora may do just as well or even better in some cases. Increasing the dimension size of embeddings after a point leads to poor quality or performance. In addition, using a relatively small corpus, we obtain better WordSim scores, corresponding Spearman correlation, and better downstream performances (with significance tests) compared to the original model, which is trained on a 100 billion-word corpus.

Open Access March 30, 2022

Post-quantum cryptography-driven security framework for cloud computing

Henry Chima Ukwuoma, Gabriel Arome, Aderonke Thompson, Boniface Kayode Alese

Page range: 142-153

More Download PDF

Abstract

Data security in the cloud has been a major issue since the inception and adoption of cloud computing. Various frameworks have been proposed, and yet data breach prevails. With encryption being the dominant method of cloud data security, the advent of quantum computing implies an urgent need to proffer a model that will provide adequate data security for both classical and quantum computing. Thus, most cryptosystems will be rendered susceptible and obsolete, though some cryptosystems will stand the test of quantum computing. The article proposes a model that comprises the application of a variant of McEliece cryptosystem, which has been tipped to replace Rivest–Shamir–Adleman (RSA) in the quantum computing era to secure access control data and the application of a variant of N-th degree truncated polynomial ring units (NTRU) cryptosystem to secure cloud user data. The simulation of the proposed McEliece algorithm showed that the algorithm has a better time complexity than the existing McEliece cryptosystem. Furthermore, the novel tweaking of parameters S and P further improves the security of the proposed algorithms. More so, the simulation of the proposed NTRU algorithm revealed that the existing NTRU cryptosystem had a superior time complexity when juxtaposed with the proposed NTRU cryptosystem.

Open Access May 16, 2022

BiSHM: Evidence detection and preservation model for cloud forensics

Prasad Purnaye, Vrushali Kulkarni

Page range: 154-170

More Download PDF

Abstract

The cloud market is growing every day. So are cloud crimes. To investigate crimes that happen in a cloud environment, an investigation is carried out adhering to the court of law. Forensics investigations require evidence from the cloud. Evidence acquisition in the cloud requires formidable efforts because of physical inaccessibility and the lack of cloud forensics tools. Time is very crucial in any forensic investigation. If the evidence is preserved before the cloud forensic investigation, it can give the investigators a head start. To identify and preserve such potential evidence in the cloud, we propose a system with an artificial intelligence (AI)-based agent, equipped for binary classification that monitors and profiles the virtual machine (VM) from hypervisor level activities. The proposed system classifies and preserves evidence data generated in the cloud. The evidence repository module of the system uses a novel blockchain model approach to maintain the data provenance. The proposed system works at the hypervisor level, which makes it robust for anti-forensics techniques in the cloud. The proposed system identifies potential evidence reducing the effective storage space requirement of the evidence repository. Data provenance incorporated in the proposed system reduces trust dependencies on the cloud service provider (CSP).

Open Access May 23, 2022

Two hide-search games with rapid strategies for multiple parallel searches

Peter E. Creasey

Page range: 171-180

More Download PDF

Abstract

Making a rapid unpredictable decision from N N choices of unequal value is a common control task. When the cost of predictability can be modelled as a penalty hidden under a single option by an intelligent adversary, then an optimal strategy can be found efficiently in O ( N log N ) O\left(N\log N) steps using an approach described by Sakaguchi for a zero-sum hide-search game. In this work, we extend this to two games with multiple parallel predictions, either coordinated or drawn independently from the optimal distribution, both of which can be solved with the same scaling. An open-source code is provided online at https://github.com/pec27/rams.

Open Access May 23, 2022

A student-based central exam scheduling model using A* algorithm

Mehmet Sinan Başar, Sinan Kul

Page range: 181-190

More Download PDF

Abstract

In this study, a student-based placement model using the A* algorithm is proposed and applied to solve the problem of placing the courses in exam sessions. The application area of the model is midterm and final exams, conducted by the Open Education Faculty. The reason for choosing open education exams for the practice is that the exams are applied across the country and more than 100,000 students participate. The main problem is to obtain a suitable distribution that can satisfy many constraints simultaneously. In the current system, the lessons in the sessions were placed once using the curriculum knowledge. This placement plan is applied in all exams. When the placement is done according to the curriculum information, the courses in the sessions cannot be placed effectively and efficiently due to a large number of common courses and the large number of students taking the exam. This makes the booklets more expensive and the organization more prone to errors. Both the opening of new programs and the increase in the number of students regularly lead to the necessity of placing the classes in sessions dynamically each semester. In addition, to prevent conflicts with the calendars of other central exams, it is necessary to conduct all exams in three sessions. A better solution was obtained by using a different model than the currently used model in the study. With this solution, distribution of the courses of successful students with few courses to all sessions is provided, and difficult courses of unsuccessful students who have a large number of courses were gathered in the same session. This study can support future studies on two issues: the first issue is the approach of using the course that will be taken by most students instead of the courses taught in most departments in the selection of the course to be placed in the booklet. The second issue is to try to find the most suitable solution by performing performance tests on many algorithms whose performance has been determined by many academic studies.

Open Access May 30, 2022

Deep learning-based ensemble model for brain tumor segmentation using multi-parametric MR scans

Suchismita Das, Srijib Bose, Gopal Krishna Nayak, Sanjay Saxena

Page range: 211-226

More Download PDF

Abstract

Glioma is a type of fast-growing brain tumor in which the shape, size, and location of the tumor vary from patient to patient. Manual extraction of a region of interest (tumor) with the help of a radiologist is a very difficult and time-consuming task. To overcome this problem, we proposed a fully automated deep learning-based ensemble method of brain tumor segmentation on four different 3D multimodal magnetic resonance imaging (MRI) scans. The segmentation is performed by three most efficient encoder–decoder deep models for segmentation and their results are measured through the well-known segmentation metrics. Then, a statistical analysis of the models was performed and an ensemble model is designed by considering the highest Matthews correlation coefficient using a particular MRI modality. There are two main contributions of the article: first the detailed comparison of the three models, and second proposing an ensemble model by combining the three models based on their segmentation accuracy. The model is evaluated using the brain tumor segmentation (BraTS) 2017 dataset and the F 1 score of the final combined model is found to be 0.92, 0.95, 0.93, and 0.84 for whole tumor, core, enhancing tumor, and edema sub-tumor, respectively. Experimental results show that the model outperforms the state of the art.

Review Article

Open Access May 25, 2022

Sign language identification and recognition: A comparative study

Ahmed Sultan, Walied Makram, Mohammed Kayed, Abdelmaged Amin Ali

Page range: 191-210

More Download PDF

Abstract

Sign Language (SL) is the main language for handicapped and disabled people. Each country has its own SL that is different from other countries. Each sign in a language is represented with variant hand gestures, body movements, and facial expressions. Researchers in this field aim to remove any obstacles that prevent the communication with deaf people by replacing all device-based techniques with vision-based techniques using Artificial Intelligence (AI) and Deep Learning. This article highlights two main SL processing tasks: Sign Language Recognition (SLR) and Sign Language Identification (SLID). The latter task is targeted to identify the signer language, while the former is aimed to translate the signer conversation into tokens (signs). The article addresses the most common datasets used in the literature for the two tasks (static and dynamic datasets that are collected from different corpora) with different contents including numerical, alphabets, words, and sentences from different SLs. It also discusses the devices required to build these datasets, as well as the different preprocessing steps applied before training and testing. The article compares the different approaches and techniques applied on these datasets. It discusses both the vision-based and the data-gloves-based approaches, aiming to analyze and focus on main methods used in vision-based approaches such as hybrid methods and deep learning algorithms. Furthermore, the article presents a graphical depiction and a tabular representation of various SLR approaches.

About this journal

SPECIAL ISSUES

Special Issue based on the 2022 IEEE 16th International Scientific Conference on Informatics

Guest Editors:
Valerie Novitzká, Technical University of Kosice, Slovakia
William Steingartner, Technical University of Kosice, Slovakia
Deadline for submissions: March 31, 2023

Special Issue on Programming Models and Algorithms for Big Data

Guest Editors:
Fadi Al-Turjman, Faculty of Engineering, Near East University, Turkey
Shahid Mumtaz, Instituto de Telecomunicações Aveiro, Aveiro, Portugal
Deadline for submissions: January 17, 2023

Open Computer Science is a fully peer-reviewed, open access, electronic-only journal publishing works of wide significance, originality and relevance. The journal provides the readers with free, instant, and permanent access to all content worldwide; and the authors with extensive promotion of published articles, long-term preservation, no space constraints and quick publication.

Open Computer Science is listed inter alia by Web of Science (ESCI - Emerging Sources Citation Index) and SCOPUS. Our standard policy requires each paper to be reviewed by at least two Referees and the peer-review process is single-blind.

Open Computer Science publishes original and significant research results in the following fields:

algorithms and complexity theory
artificial intelligence
bioinformatics
networking and security systems
programming languages
system and software engineering
theoretical foundations of computer science

The journal considers submissions of different types of articles, including "Research Articles" as well as “Review Articles”, which present the existing literature and achievements on the specific topic from new perspectives. We accept both theoretical and empirical contributions.

Why read

Open Computer Science is a premier source of a high quality research. Thanks to Open Access all articles are freely available to the academic community worldwide.