These articles have been peer-reviewed and accepted for publication but are pending final changes. The articles will be published in near future as soon as they are assigned to the respective issues.
A Novel Text Mining Approach to Securities and Financial Fraud Detection of Case Suspects
SUNDAR KRISHNAN, Narasimha Shashidhar, Cihan Varol, ABM Rezbaul Islam
Securities or stock fraud is a type of financial fraud involving securities or asset markets that can result in criminal charges and jail time. Detecting securities fraud from a pile of electronic evidence without automation, statistical methods, and analytics can be a mammoth exercise for investigation teams due to the ever-increasing volumes of electronic data as case evidence. In this study, the authors propose a machine learning and neural network-based platform consisting of various analytical approaches that can assist financial/securities/accounting forensic investigators, legal teams, paralegals, digital forensic investigators, and auditors in financial fraud case investigations such as “insider trading fraud” and “pump and dump fraud”. This platform can help reduce investigation time and increase efficiency in identifying internal trading fraud and pump and dump fraud indicators.
KEYWORDS - Supervised Learning, Digital Forensics, Forensic Accounting
A Novel Text Mining Approach to Sexual Harassment Detection of Case Suspects
SUNDAR KRISHNAN, Narasimha Shashidhar, Cihan Varol, ABM Rezbaul Islam
Sexual harassment cases often go unreported and can be difficult for an investigator to detect when working with large volumes of digital evidence of an investigation. Artificial Intelligence can be a promising solution to help identify instances of sexual harassment, especially from written communication. In this research, an approach to detect indicators of sexual harassment is proposed using supervised and unsupervised learning coupled with the application of Bidirectional Encoder Representations from Transformers (BERT) and Snips NLU. The models are then applied against synthetic digital forensic evidence data for detection of sexual harassment indicators from textual digital evidence.
KEYWORDS - Sexual Harassment, Digital Forensics, Ediscovery
Session Initiation Protocol: Security Issues Overview
Bruno Cruz, Rui Filipe Pereira
The leading method of correspondence is clearly through voice trade. There are essentially two different ways through which voice can be effortlessly communicated on an organization: PSTN (Public Switched Telephone Network) and VoIP (Voice over Internet Protocol).

Mainly represented by SIP, VoIP protocols and implementations contain several vulnerabilities, particularly related to their complexities and in the face of interoperability of telephony equipments.

It was by identifying a lack of literature with focus in security and potential vulnerabilities of the SIP Protocol that we propose in this document. We attempt to provide a theoretical analysis from security aspects used by one of the signaling call protocols, Session Initiation Protocol (SIP).

It is intended to lucidly illustrate and identify threats, vulnerabilities, security mechanisms, developed methods and protocols and, finally over time improvements.
KEYWORDS - Session Initiation Protocol (SIP), SIP Security, Voice over IP (VoIP).
How has Entrepreneurship Opportunity Formation amongst immigrants been influenced by the Covid-19 pandemic?
Osa-Godwin Osaghae, Thomas. M. Cooney
A body of evidence exists which suggests that stable economic conditions support entrepreneurial opportunity formation within a national context. However, despite the advent of the COVID-19 Pandemic and recent global economic uncertainty, entrepreneurial activity is continuing to flourish across the globe. This article explores the possible factors driving entrepreneurship opportunity formation in the current climate. The study employed a comparative narrative analysis of literature relating to entrepreneurship opportunity formation, environmental change (e.g. COVID-19 pandemic) and demand expansion (e.g. market extension resulting from environmental changes). The triangulation of the literature from these diverse topics leads to the conclusion that environmental changes and irregularity creates demand expansion that can drive entrepreneurship in any climate. This article contributes to knowledge by suggesting that irregular events within the environment (not stable economic growth alone) can positively influence entrepreneurship opportunity formation.
KEYWORDS - Environment, climate change, entrepreneurship opportunity
Faith and Management at Theological Seminaries
Johannes Schroeder
Leadership for theological training institutions has been subject to the influences of the academization of the organizations and a concurrent professionalization of the leader’s role. The doctoral thesis “Leading evangelical seminaries in German-speaking Europe: A transcendental phenomenology” (Schroder, 2016) is an empirical phenomenology among lead administrators of protestant seminaries in the free-church context. It describes the participants’ experiences in their lived phenomenon of leadership within the theoretical frameworks of spiritual leadership, servant leadership, and workplace spirituality. Six lead administrators voice their experiences regarding their management responsibilities and their spirituality. This article will summarize the study while focusing on findings and themes on issues of faith and management. As the most profound theme emerged the leader’s spirituality as the primary source of motivation and meaning for the tasks and experiences contained within their leadership role.
KEYWORDS - Spiritual Leadership, Servant Leadership, Workplace Spirituality
"Right-sized", Contextualized Faith: The Challenging Cooperation of FBDOs with Institutional Donors
Matthias Hoehne
This paper explores the challenging cooperation between faith-based development organizations (FBDOs) and institutional donors. It argues that FBDOs need to find their right-sized, contextualized faith by understanding their unique developmental and spiritual contributions and the challenges they face. Five distinctive contributions of FBDOs are described: extensive networks, community engagement at multiple levels, holistic and innovative approaches, higher trust level, and spiritual capital. Then five challenges they face are set forth: instrumentalization of FBDOs, a need constantly to prove themselves, transparency, proselytism, and downplay of spiritual identity. In light of the FBDOs distinctive contributions and challenges, three practical recommendations for FBDOs to achieve right-sized, contextualized faith are presented, namely to aim for a) right-sized, contextualized programs, b) right-sized funding, and c) right-sized faith.
KEYWORDS - FBDO, faith-based organizations, institutional donors, faith identity, development, spirituality
Ethical theory vs. practical worl-life: implications for organizations
Ana Maria Cabodevila
In business, the relationship between ethical theory and a practical work-life is challenging, since there is a strong ethics-practice contrast: work as the practical dimension and ethics as the theoretical dimension. There is, thus, a gap between theory and practice, and the imbalance is obvious in a majority of organizations, be they profit or non-profit. How do organizations respond to ethical theories of business? On the one hand, ethical convictions and practices may be understood as being merely subsets of national culture, and so are thus not, or only partially, considered in the workplace. On the other hand, organizations may perceive business ethics as being rather limiting with regard to practice owing to an abstract understanding of ethics (Brügger & Kretzschmar, 2015, p. 3). Business ethics theory is, thus, often perceived as being difficult to put into practice. The purpose of this article is to show that for many centuries, theory and practice, or faith and working life, were not separated in the way that they apparently are today. The aim is to use the examples shown to encourage organizations and responsible persons to break down the artificial divisions introduced by the Enlightenment in such a way that ethical thinking in working life no longer remains a foreign concept but can indeed find its way into daily work.
KEYWORDS - Business ethics, Enlightenment, NGO
On the reciprocal relationship between faith and management
Volker Kessler
This article describes the complex reciprocal relationship between faith and management. Firstly, faith-based organizations have to be managed. Some management methods will foster the faith; others will affect the faith negatively. Each faith group will have to look for management methods suited to the specific faith of that group. Secondly, faith has an influence on management. This can happen implicitly or by intention. An example of the latter is when managers just copy concepts from a faith group because of their success, without necessarily sharing their faith. Or it happens when believers want to implement the standards of their faith at work. The concept faith@work can be problematic if it is a single-faith approach within a secular work environment because it might lead to injustice. Due to the reciprocal relationship between faith and management, we can discover the re-entry of religious terms or concepts: these terms originate in Christian faith, enter the management sciences and from there re-enter the Christian faith. The examples “servant leadership” and “vision” show the subtle change of meaning that occurs when words wander between the two worlds, thus becoming “false friends” to the faith group.
KEYWORDS - Economic theology, Faith-based organizations, faith@work
Production Portfolio of Christian Charities - Doing the Right Thing and Keeping the Mission
Steffen Flessa
Christian charities are non-profit organizations (NPO) built on the Christian virtue of love. However, being a Christian NPO does not automatically ensure that Christian management is practiced. In this paper, the production portfolio is seen as the main criterion to determine whether a Christian NPO is indeed following Christian principles in its management. A flow chart is developed to define four categories of product lines: Touchstones are services with high relevance to the mission of the organization but without sufficient funding. Stars also make an important contribution to the achievement of the target system and are well-financed. Cash cows receive sufficient funding as well, but are not (anymore) relevant for the achievement of the original mission of the organization. Goiters might have been touchstones, stars and cash cows in former times, but today they neither do fulfil the mission of the organization, nor do they produce a positive margin. Christian charities should develop touchstones by constantly seeking upcoming existential needs of people and developing respective services. However, there is a risk that older and growing Christian charities lose their calling. Based on the Greiner curve we argue that the management of these organizations must motivate and coach their staff so that they remain dedicated to the original mission while the organization faces severe crises. By a quick and comprehensive analysis of the production program and support for the staff in particular during transgression phases, Christian management in Christian charities becomes a reality.
KEYWORDS - Christian management, Nonprofit Organization, Production portfolio
Network governance and co-production of community services. The interregional Italian - Swiss program D.E.A. (SIBRM9)
Veronica Giuliani, Davide Maggi, Cinzia Zugolaro
Over the last decades, the implementation of public networks to address social issues has become more the rule than the exception. Especially in the public sector, the interactions between different actors, and thus the activation of networks, ought to find shared solutions for problems of general interest that neither government nor other spheres of society can meet on their own. Using participative approaches, public networks reshape the traditional redistributive welfare paradigms and responsibilities in the production of social wellbeing. This paper studies the structure, roles, and performance of the network built to implement the interregional Italian-Swiss project D.E.A. (Diversità E Arti performative per una società inclusiva del terzo millennio), a European program that involves public administration, non-profit organization, universities and citizens, to promote and support the participation in civic activities of fragile people at risk of social exclusion. Using a combined approach this paper analyzes the case study both from analytical and governance perspectives: it proposes a unified theoretical framework based on the extant literature, to explore the nature and the related critical factors underlying the success of the public network under study.
KEYWORDS - network, governance , participation
Assessing the key barriers to innovation acceptance in Iranian construction companies
Hadi Sarvari, Daniel W.M. Chan, Reza Soltani
The emergence of innovation in recent decades has attracted the attention of major companies around the world to its productivity and importance. As a developing country, Iran has great potential in the construction industry, but due to the lack of a proper strategy and the lack of a favorable process in innovation, growth and development has not been appropriate. The purpose of this study is to identify and prioritize the key barriers to innovation in Iranian construction companies, which have been answered by 62 board members, consultants, contractors, supervisor and computational engineers, experts and suppliers in the construction industry. About 60 percent of participants are from the private sector, 29 percent from the public sector, and 11 percent from both sectors. Most of the participants were between 25 and 45 years old and the majority had a master's and bachelor's degree and had a experience of 5 to 20 years. In the questionnaire, the main factors of research, including: cultural, financial, organizational, and managerial, had the most effect, and human, investigate, and research factors had a moderate effect, and external factors had the least effect. The results showed that the main determinants of non-use of innovation in construction companies can be classified into internal and external factors. The factors was evaluated based on the 5-point Likert scale measurement. The results shows that all factors identified as barriers to innovation acceptance in Iranian construction companies.
KEYWORDS - Innovation, Barriers, Construction
Five Lessons by C.S. Lewis' Narnia, Discovered for Agile Teams
Emanuel Kessler
People could learn from stories. In this article, the question is asked: What can people working agile learn from “The Chronicles of Narnia”? With this goal in mind, five scenes are explained and the lessons are worked out. Since there are a large number of ways of working agile, this article is focused on the Kanban Method and Scrum as agile ideas.
KEYWORDS - C.S. Lewis, Narnia, agile, Kanban Method, Scrum
The Importance of Stakeholder Collaboration and Co-creation in the Social Innovation Creation Process
Emmanuel Osigwe
Over the years social innovation (SI) has been offering useful solutions to social problems in societies. SIs are created by social enterprises (SEs) that may operate as a non-profit, for-profit, or hybrid enterprise with the primary mission to create social impact to benefit societies. The literature on SI is emerging, thus greater attention is given to understanding and conceptualising SI. Recent studies have shown that the process of SI largely remains unexplored. As such, little is known about the importance and the role of external actors in the process. This study examines the importance of stakeholder co-creation within the SI ecosystem through inductive analysis of interview data from 20 SEs in four countries. The study finds SE seek strategic collaborations, engage critical beneficiaries, and explore social networks external to their enterprise when seeking to co-create SI.
KEYWORDS - Social innovation, Stakeholder engagement, Collaboration
Tongesai Mpofu, Shephard Makurumidze
The purpose of the paper is to find how SMEs in Zimbabwe managed to implement change with respect to COVID-19. The pandemic has changed the way of doing things in the entire world. To survive in situations, organisations need to change. All the SMEs in Zimbabwe may be trying their level best to implement change successfully however, all of them are facing severe issues. The purpose of this study is to identify the major hurdles in implementing change and identifying a way out to cope with the situation based on available literature on SMEs and change management. The other purpose is to find the challenges faced by these SMEs in managing and implementing change.

This study follows a synthesized literature review methodology and is a kind of review paper. In this research, literature that has been written on crisis management and especially over the COVID-19 has been reviewed. In this review analysis, previous studies regarding change management and the importance of SMEs in economic sustainability have been discussed. The paper is a review of existing literature and will identify how the organizations can survive through implementing change successfully.

This study found out that SMEs in Zimbabwe have being struggling to manage and implement change due to non-availability of financial resources. This study is limited to SMEs in Harare province only it does not cover all SMEs in Zimbabwe. The study is significant for SMEs in the entire world. The study is useful for understanding the issues that are related to managing and implementing change in small and medium enterprises especially while facing contingencies. The findings of the study and the literature review led to the conclusion that without implementing change successfully, it is impossible to survive during the COVID-19 pandemic. In order to deal with the crisis of resistance to change organisations should motivate and train its employees. The study recommends SMEs in Zimbabwe to manage change by training and motivating employees and the government to support the SMEs sector in order for it to survive since its a backbone of the Zimbabwean economy.
KEYWORDS - Change Management, Implementing Change, Small and Medium Enterprises, COVID-19, Performanc
Tongesai Mpofu, Shephard Makurumidze
COVID-19 pandemic brought a lot of challenges for the global community from the year 2020. The spread of this virus led to unparalleled health crisis in all the countries across the world. The COVID-19 pandemic has also caused unprecedented panic and disruptions for both the public and private sectors and is considered an experiential threat to the global economy with governments and businesses grappling with the effects. In Zimbabwe, the pandemic has caused challenges and opportunities for Small and Medium Enterprises (SMEs) and in some cases threatening their own survival and forcing many SMEs to change their focus in order to manage the crisis. Some SMEs have adopted crisis management and alternative business response efforts. The main objectives of this paper were to determine the challenges of COVID-19 pandemic to SMEs in Zimbabwe and identify the emerging opportunities arising as a result of the pandemic. The methodology of this article is secondary information where various literature on challenges and opportunities for small to medium businesses was analyzed. Secondary data results show that many small businesses are suffering and the COVID-19 caused destruction for many small to medium businesses. It is difficult to survive with low income, jobs were lost and frail marketing performance. Findings from this study show that amongst the crisis caused by COVID-19, numerous opportunities have also emerged for innovative Zimbabwean entrepreneurs to explore. The firms should reduce expenses, adapt to new technology in order to be competitive and survive and utilise other business opportunities that have been opened up by COVID-19 pandemic. The study recommends that SMEs should first assess the damages of the pandemic to the businesses and then find the strategies to mitigate the negative effects of the pandemic and maximize the new opportunities. Furthermore, the study recommends the companies to adopt digital marketing, and use different alternatives to deliver their products and recover from crisis.
KEYWORDS - COVID-19, Pandemic, SMEs, Challenges, Emerging Opportunities, Digital
The Social Responsibility and the Studies about the Police Forces: An Analytical Review
Mónica de Melo Freitas, Ivone Freire e Costa, Marco Meneguzzo, Rocco Frondizi
This study aimed to verify how Social Responsibility in polices is being discussed by authors from different scientific areas. The main goal means to identify the ethical values and the practices invoked by the authors to justify the phenomena of social responsibility in police sector. Our initial hypothesis is that the authors of police studies are approaching it as merely a result of the implementation of models and practices typical of the privative sector into the public sector through the New Public Management. During our study we had found 2.010 papers that discussed the Social Responsibility in public sector but only 18 tried to understand it in police forces applying authors from Law, Criminal and Justice, Social and Behavioral Sciences, Economy, Business and Accountability Sciences and Neuro- Cognitive Sciences and comprehensive models of analyzing. The findings produced by the study showed that Social Responsibility is an unexplored theme until today in police studies. It highlighted also that the New Public Management theories offer an interesting framework theory to understand the Social Responsibility phenomena within police forces. Whatever, it enforces the necessity to understand the Social Responsibility in institutional and cognitive ways because authors appointed that their personal values conduct the modes in which normative orientations are understood and applied by them. This paper is innovative because offers new insights about the Social Responsibility within police forces, but it did not produce inputs that allow characterize the performance of the police forces in any country.
KEYWORDS - Social Responsibility , New Public Management , Police Forces
Sustainability: Future Orientation Through Engagement of MSMEs
Deepti Prakash, Parul Manchanda, Twinkle Arora
Purpose: Enterprises have been facing a concern from the government to take an initiative in the various environmental maintenance and in the implementation of various sustainable practices laterally with the satisfaction of the customer demands. Thus, the enterprises are enforced to implement sustainable ways of doing business, which would help them achieve competitive advantage in the long run. This paper intends to institutionalise the various sustainability measures (through the leadership approaches and the theoretical approaches) in the various Micro, Small and Medium Enterprises (MSMEs).

Design/methodology/approach: This research paper is a general review for highlighting the varied reasons and unreasons behind the various enterprises practising sustainability initiatives in the real business scenario. A thorough and wide exploratory search was made from the existing literature with the help of online databases. The results are presented in the form of descriptive findings.

Findings: The research paper concludes that, sustainability is not a unitary concept, but involves a throng of efforts (to explain, the activities, actors and the resources employed). The MSMEs require explicit thought, in case of business plans for sustainability as it is by one way or another not equivalent for the large firms. It has also been brought about that the MSMEs require a different way to support sustainability in comparison to the various large organisationswhich hold a varied set profile and resources.

Practical implications: Sustainability, today is a concern for everybody in the civilisation, this is because of the changes in the climate that have been observed and the growing problem of global warming. This research work, may enable the MSME managers to reconsider the whole business strategy, and making sustainability as an important inclusive element of the same, and thus practising it too.

Research Limitations: Sustainability, has been an important concern to the society in general which points out that there can be plenty opportunities for various organisations to identify strategies that will have a bearing and may positively advance the social and environmental performance. However, this research work, does not provide an empirical evidence and support but offers insights on engaging MSMEs in sustainability.

Originality/ Value: This research contributes to the area of literature by providing a review, for the various considerations and occasionsfor the various business strategies for sustainable development and its varied applications to the certainties of business operations in various MSME enterprises.
KEYWORDS - MSME, Small and Medium Enterprises, Sustainable Development, Leadership Approaches, Business Sustainability, Environmental Management Practices.
Unicode-based Data Processing for Text Classification
Akash Sedai, Ben Houghton
In this paper we demonstrate a Unicode based text data processing ap- proach for machine learning classification. The fields are first converted to Uni- code, and then the features are generated by splitting the characters by vowels contained within each field. The fields are labelled into classes. The model outputs the class predictions for each field.
KEYWORDS - text classification, text processing, NLP
Suffix-stripping Algorithms and Transducers for the Fulani Language
ZOULEIHA ALHADJI IBRAHIMA, Dayang Paul, Kolyang, Guidana Gazawa Frederic
Because to the large and constantly increasing amount of information available on the Internet, users are facing with challenges and difficulties to satisfy their need. In fact, the objective of today's information retrieval systems is no longer accessing to information but the search and filtering of relevant information. The language used for searching information plays a major role. If we consider resource scarce local or national languages, the situation becomes even more challenging. Many African languages fall into the group of resource scarce languages. Therefore, there is a need to explore and build more specialised information systems that enable speakers of African languages to discover valuable information across linguistic and cultural barriers. As one of the most dispersed languages in Africa, the Peul also called Fulani language suffers from a significant handicap in its computerisation and automatic processing due to the inexistence of digital and linguistic resources. Considering the fact that a devoted care and attention to conserve, guarantee sustainability of languages is important, few studies and computerisation works have been carried out on African Languages such as Fulani. The aim of this work is to lay some bricks towards tools for the automatic processing of the Fulani language. This language belongs to several dialectal areas and there are almost no digital documents of the Fulani language of the Adamaoua dialectal area. The originality of this work is first the digitization of Noye Dominique Fulani dictionary from North Cameroon; we then studied stemming approaches for Fulani words using transducers that clearly show how to remove classifiers from words in order to obtain the stem. To date, no research work has been done in this direction for the Fulani language or for native African languages similar to Fulani. This approach is crucial in information retrieval systems because it allows translation and classification of documents as well as indexing of words. To specify the stemming approaches, we have adapted the stemming algorithms of Lovins and Porter to the Peul language, knowing that they are the best known and have the advantage of being applied to other languages. Finally, the evaluation of these stemming methods was done using the method of Christ Paice. Based on the principle that words sharing the same stem are likely to share a unity of meaning, we undertook a morphological analysis of 5186 Fulani words.
KEYWORDS - Peul, Fulani, Suffix-stripping, Stemming, Linguistic, Transducers.
The research is aimed at developing a real time intruder detection system using facial recognition and deployable on automated teller machines in Nigerian Banking sector. This research was implemented using python programming language and raspberry pi device. Haar cascade classifiers was used to implement face detection and recognition. A total of six thousand images were used to train the classifier. Five thousand positive images and one thousand negative images. The performance of the result was evaluated using five major metrics; Sensitivity, specificity, accuracy, precision and processing time. The results obtained have sensitivity 90.6%, specificity 94%, precision 94.1%, accuracy 92.2%, a maximum of 16 seconds was taken to detect a face and a maximum of 45 seconds was recorded to recognize a detected face. Achieving a maximum of 45 seconds with 92.2% accuracy real time, indicates this work can be deployed on any real-life system.
KEYWORDS - Image recognition., face detection, raspberry pi
Performance evaluation of Denial of Service Attack Detection on SIP based VoIP.
Abdirisaq, Othman O. Khalifa, Nantha Kumar Subramaniam
Recent trends have revealed that SIP based IP-PBX DoS attacks contribute to most overall IP-PBX attacks which is resulting in loss of revenues and quality of service in telecommunication providers. IP-PBX face challenges in detecting and mitigating malicious traffic. In this research, Support Vector Machine (SVM) machine learning detection & prevention algorithm were developed to detect this type of attacks Two other techniques were benchmarked decision tree and Naïve Bayes. The training phase of the machine learning algorithm used proposed real-time training datasets benchmarked with two training datasets from CICIDS and NSL-KDD. Proposed real-time training dataset for SVM algorithm achieved highest detection rate of 99.13% while decision tree and Naïve Bayes has 93.28% & 86.41% of attack detection rate, respectively. For CICIDS dataset, SVM algorithm achieved highest detection rate of 76.47% while decision tree and Naïve Bayes has 63.71% & 41.58% of detection rate, respectively. Using NSL-KDD training dataset, SVM achieved 65.17%, while decision tree and Naïve Bayes has 51.96% & 38.26% of detection rate, respectively. The time taken by the algorithms to classify the attack is very important. SVM gives less time (2.9 minutes) for detecting attacks while decision tree and naïve Bayes gives 13.6 minutes 26.2 minutes, respectively. Proposed SVM algorithm achieved the lowest false negative value of (87 messages) while decision table and Naïve Bayes achieved false negative messages of 672 and 1359, respectively.
KEYWORDS - Voice over IP, , Session Initiation Protocol, Denial of Service
Performance Evaluation and Analysis of Supervised Machine Learning Algorithms for Bitcoin Cryptocurrency Price Forecast
Mohammed Ajuji , Sani Abba, Souley Boukhari, Amina Nuhu Muhammad
Earlier to the advent of computers and the internet. Transactions such as buying, selling, hiring, and cash transfer are performed physically, hand-to-hand and/or face-to-face using hard-printed currency also known as traditional means. The recent advances in internet and networking technologies have significantly refurbished and improved the methods and limitations of the traditional ways, through cryptocurrency or digital money especially in terms of cost, speed, and access. These technologies which bring people together irrespective of geographical location have fashioned a revolution in trading and transaction processing; online transaction processing and real-time processing. However, like every other pioneering development, this is not without resistance from stakeholders, whom have been using the traditional means for long; its validity and legitimacy have been seriously challenged. In this study, several models leveraged to forecast bitcoin price were Linear Regression (LR), K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Decision Tree (DT), Elastic Net Regression (EN), Lasso Regression (Lasso) and Ridge regression (RR). The models’ accuracies were determined and evaluated using Mean Absolute Error (MAE), Mean Square Error (MSE), and R Square Error (R2). It revealed good performance except for SVM which falls in the negative even after fine-tuning and improved performance. The LR led in performance, then EN, Lasso, and RR. Decision Tree on the other hand present an encouraging and challenging result. Whereas the SVM model presents worst-case prediction accuracy of -22.38%. Therefore, the linear regression model has the best fit for bitcoin price prediction amongst the algorithms.
KEYWORDS - Bitcoin, Cryptocurrency, Machine learning
A Hybrid Face Recognition Method based on Face Feature Descriptors and Support Vector Machine Classifier
Rafika Harrabi Harrabi
Face recognition is a technique used to identify/verify human identity based on their facial features. A technique allows, based on facial features to authenticate / identify a person. However, for human identification or identity authentication based on face recognition technology, the appropriate determination of the face features plays a crucial role, since the identification of the Human is given directly by the classification of these characteristics.

In this paper, we propose a new face recognition method based on face feature descriptors and Support Vector Machine (SVM) algorithm. The face feature descriptors are used to extract and select the statistical features, whereas, the SVM algorithm is employed to classify the different features and to obtain optimal Human face recognition.

The feature extraction step is the major phase of the recognition cycle. It is employed to extract the features for any human face located in the first step. The accomplishment of this step controls the success of subsequent steps. For that, the main objective of this work is to determine of the best method of feature extraction.

To do the indexation of persons face, the Histogram of Oriented Gradient features (HOG), Gabor features and Discrete Cosine Transform features (DCT) are employed to extract the feature vectors for any human face.

In addition, the face recognition method, proposed in this paper, is conceptually different and explores a new strategy. In fact, instead of considering an existing face recognition procedure, the proposed technique rather explores the benefit of combining several approaches.This method is a hybrid face recognition technique, which integrates both the results of the HOG, and the SVM technique, in which the HOG method is used as the initial seed for the classification procedure.

Experimental results from the proposed method are validated and the face recognition rate for the ''ORL'' and cropped ''Yale B'' datasets is evaluated, and then a comparative study versus existing techniques is presented. The highest face recognition rate of the used dataset is obtained by the proposed method. In addition, the use of the proposed HOG_SVM method to build face recognition systems can achieve excellent results when the dataset size is large, and therefore it can be used in different security and authentication systems.
KEYWORDS - Face Recognition, Feature Extraction, HOG, SVM, DCT, Gabor, Classification.
MVC Architecture from Maintenance Quality Attributes Perspective
Safia Nahhas
This paper provides an explanatory study on MVC (Model-View-Controller) architecture from the perspective of maintenance. It aims to answer a knowledge question about how MVC architecture supports the maintainability quality attributes. This knowledge boosts the potential of utilizing the maintainability of MVC from several sides. To fulfill this purpose, we investigate the main mechanism of MVC with focusing on maintainability quality attributes. Accordingly, we form and discuss MMERFT maintainability set that consists of Modifiability, Modularity, Extensibility, Reusability, Flexibility, and Testability. Besides investigating the mechanism of MVC regarding MMERFT quality attributes, we explain how MVC supports maintainability by examining measures and approaches such as: complexity of code by using a cyclomatic approach, re-engineering process, use of components, time needed to detect bugs, number of code lines, parallel maintenance, automation, massive assignment, and others. Therefore, this paper is dedicated to providing a concrete view of how MVC gets along with maintainability aspects in general and its several attributes particularly. This view helps to maximize the opportunity of taking advantage of MVC's maintainability features that can encourage reconsidering the maintenance decisions and the corresponding estimated cost. The study focuses on maintainability since software that has high maintainability will have the opportunity to evolve, and consequently, it will have a longer life. Our study shows that MVC generally supports maintainability and its attributes, and it is a recommended choice when maintenance is a priority.
KEYWORDS - MVC Architecture, Maintainability, Modifiability, Modularity, Extensibility, Reusability, Flexibility, Testability.
Product Sentiment Analysis for Amazon Reviews
Arwa M. AlQahtani
Recently, eCommerce has witnessed rapid development. As a result, online purchasing has grown, and that has led to growth in online customer reviews of products. The implied opinions in customer reviews have a massive influence on customer's decision purchasing, since the customer's opinion about the product is influenced by other consumers' recommendations or complaints. Accordingly, product purchases could increase or not based on consumers' reviews. Also, customer opinions could help firms improve insights into customer interests, hence improve their products or services. This research provides an analysis of the Amazon reviews dataset and studies and sentiment classification with machine learning approaches. First, the reviews were transformed into vector representation using different techniques, i.e., bag-of-words, TF-IDF, and GloVe. Then, we trained various machine learning algorithms, i.e., Logistic Regression, Random Forest, Naïve Bayes, Bidirectional Long-Short Term Memory, and BERT. After that, we evaluated the models using accuracy, F1-score, precision, recall, and cross-entropy loss function. The experiment was conducted on multiclass classifications, then we selected the best performing model and re-trained it on the binary classification.
KEYWORDS - Amazon, Data Analytics, Product Sentiment
Run-time Detection of Cross-site Scripting: A Machine-Learning Approach Using Syntactic-Tagging N-Gram Features
Nurul Atiqah Abu Talib, Kyung-Goo Doh
Ensuring the security of web applications against cross-site scripting is practically a never-ending story. With the emergence of new applications with loaded payloads of open expressiveness and versatile functionalities to provide users with interactive services, the fight is even more challenging. A new feasible approach now in growing prominence is to use machine-learning classification. In this paper, we demonstrate an approach for payload abstraction through translation of payloads into sentences of syntactic tags. This is to extract a normalized set of features of appropriate data and to minimize the problems of manually creating rules based on dangerous characteristics of payloads. We show that through abstraction and normalized features, we can accurately classify input payloads according to their proper categories. We assert that the security work becomes more sustainable by using the automaton of machine-learning technique.
KEYWORDS - Web Application Security, Cross-site Scripting, Supervised Machine-learning
Design of an Omni-Direction Robot with Spherical Wheels
Dominic Campbell, Emanuele Lindo Secco
We proposed the design of an omni-direction robot which embeds 4 spherical wheels. The wheels are connected to 4 DC motors and controlled through an L298N boards which is powered with 2 LIPO batteries. A low-cost Arduino board oversees the system by controlling the motion and speed of the motor.

The wheels of the robot integrate an omni-directional mechanism and all components have been designed in Fusion 360 (Autodesk ), manufactured with a 3D printer (Ender 5) an then assemble and integrated with the hardware and software of the system.

Preliminary tests show that the proposed solution is promising and provide a good reference for the manufacturing of low-cost robot.
KEYWORDS - Omni-directional Wheel, Osaka Wheel Mechanism, Wheeled Robotics.
Enhanced Matched Filter Theory and Applications
Kaveh Heidary
Enhanced matched filter (EMF) comprises a distortion tolerant correlation filter and its associated threshold. It is an effective signal detection tool with superior immunity to noise, distortion, and clutter, used for imagery-based authentication, classification, recognition, and tracking of targets of interest. The EMF is synthesized by combining multiple image templates of the target of interest, acquired under prescribed target states and view conditions. In autonomous vision and tracking systems, one EMF can potentially replace copious manifold of exemplar images without adversely affecting the classifier precision. This leads to proportional reduction of operation phase computational load, and concomitant smaller footprint, lighter, faster, and more power efficient smart vision systems. This paper develops the underlying theory of the EMF and provides analytical models of its operation. Filter performance results based on analytical formulations and empirical studies are presented and are compared to the performance data using virtual and real test images.
KEYWORDS - Enhanced Matched Filter, Correlation Filtering, Distortion Tolerant
A Self-Adaptive High/Low Beam Spotlight Filter in Capturing Local Structure Information for Object Contour Extraction
Roy Chaoming Hsu, Chia Hung Hsu
Contour extraction is a method in exactly obtaining an object’s contour from images. It is considered as one of the most important pre-processing for image processing applications. In this study, a self-adaptive high/low beam spotlight filter (SAHLBSF) is designed to capture local structure information for object contour extraction. The proposed SAHLBSF is inspired from the users’ experiences in car driving, where when the road is very straight and clear, a low beam light is applied, while a high beam light will be utilized when the road is winding and/or the environment is dark. Utilizing the adaptive high/low beam spotlight filter, the local structural information between two pre-selected initial contour points are captured and the candidate contour points are then determined. The spotlight filter continues for all pairs of initial points of an object such that a broadband of the object’s contour is constructed. A thinning process is then applied to obtain the final one-pixel-wide exact object contour. Experimental results using artificial and real medical images showed that better contour extraction performance can be obtained the proposed SAHLBSF than other existing methods.
KEYWORDS - Contour Extraction, Local Structure Information, Spotlight Filter
Analytics in Digital Forensics and eDiscovery Software - DevOps, Opportunities and Challenges
SUNDAR KRISHNAN, ABM Rezbaul Islam, Cihan Varol, Narasimha Shashidhar
Digital forensic and eDiscovery software have embraced analytics such as machine learning and neural networks to speed up the investigation and thereby reduce costs. Since the integrity of forensic evidence is paramount to the investigation, care should be taken when working with evidence in a analytical experiment setting. Data mined from case evidence can provide different clues and together with automation, the legal teams can better prepare legal arguments for the courtroom. In this paper, the authors develop a custom digital forensic software that leverages analytics and outline few development challenges and opportunities encountered along the way.
KEYWORDS - Digital Forensics, Ediscovery, Hybrid Learning
Sentiment Analysis of Case Suspects in Digital Forensics and Legal Analytics
SUNDAR KRISHNAN, Narasimha Shashidhar, Cihan Varol, ABM Rezbaul Islam
Sentiments of suspects in a legal case or digital forensic investigation can be of use when profiling their state of mind or feelings. Such information can help case investigators to plot their actions against case timelines, understand their instincts and build psychological profiles. In this research, the authors first assemble a fictional dataset of electronic evidence and store in a SQL Database. Next, they leverage different Machine Learning and Natural Language Processing (NLP) techniques to propose an approach to plot sentiments of case suspects linking back to the source of evidence. This information is then presented via a custom software for case investigators who can use it to pick a suspect from the case and obtain their sentiments against various electronic evidence sources that they were associated with.
KEYWORDS - Sentiment Analysis, Digital Forensics, Legal Analytics
Flare – A Community Based SOS Application
Sanil Arun Chawla
This paper presents a mobile based application that will enable to solve numerous problems with a simple solution. The project - ‘ Flare ‘ focuses on creating a community based application wherein the community helps and grows together. The application is called Flare , keeping in mind the actual flare gun that is used by like sailors or like army-men in times of need and they launch an actual flare. Similarly any user can launch a flare – a call for help.
KEYWORDS - SOS , Android , Software
Software Team Productivity Factor in Constructive Cost Model for Software Development Effort Estimation
Edward Ndarake Udo, Okure U. Obot, Peter G. Obike
One of the models used to implement software development effort estimates is the Constructive Cost Model (COCOMO) and the attributes of this model is said to contain some level of imprecision. This study was motivated by the need to estimate software development effort accurately and also reduce the imprecision contained in the COCOMO. A neuro-fuzzy constructive cost model of Kaur et al (2018) was studied and found to contain some of the desirable features of a neuro-fuzzy approach. It however handles imprecision using Adaptive Neuro-Fuzzy Inference System (ANFIS) with a large dimension of datasets and does not consider software team members productivity. This work therefore introduces software team productivity factor into the conventional COCOMO and converts it to COCOMO II using model definition manual and Rosetta Stone and also considers reducing the number of inputs from 23 to 6. With data gathered from PROMISE repository (NASA project), an ANFIS based model was built. The new model with the productivity factor was implemented along with that of Kaur et al., (2018) in MATLAB 2021 programming environment. Findings reveal that with 6 out of the 23 attributes of PROMISE datasets, the ANFIS model (Hybrid and Back Propagation) with the productivity factor performs better than the Kaur et al., (2018) model. The implication is that the productivity of the team members working on a software project can add up or reduce the actual person-hours (effort) required to develop a software. During the experiments, six (6) important COCOMO inputs that software managers should place more emphasis on during the planning stage were identified
KEYWORDS - Software Development Effort, Team Productivity Factor, Adaptive Neuro-Fuzzy Inference System
A systematic literature review on effort estimation in agile software development using machine learning techniques
Pranay Tandon, Ugrasen Suman
Agile software development is a way of frequent or continuous delivery of software. Nowadays many software industries have adopted agile for software development. The predictability and stability of traditional methods were replaced with flexibility, adaptability and agility to generate maximum value with collaboration and interaction, as quickly as possible. Effort estimation is the focused area in agile software development to achieve customer collaboration, respond to change and deliver a working software on time. Machine learning is an advanced tool to obtain effort estimation with available project data and widely used in IT industries to get accurate estimations. In this paper, we report our findings through systematic literature review that aimed at identifying the applicability, limitations and individual result of most used machine learning techniques for effort estimation in agile software development. We have also discussed suggested attributes of a robust machine learning model to achieve more accurate effort estimation.
KEYWORDS - Agile software development, Effort estimation, Machine learning
Hybrid Pair Programming in Global Software Development
Irdina Wanda Syahputri, Ridi Ferdiana
Pair programming exhibits direct communication in the same place, same time, and same context. In global software development or distributed software development, pair programming cannot be applied in full-time development because of the limited interaction, space, and time between developers. We aim to understand the pair programming pattern in global software development. We pinpoint the developer's activity based on their work habits between solo and pair activities. Furthermore, we propose hybrid pair programming to implement pair programming and solo programming in global software development. In this experiment, we construct an experiment to understand how pair programming and solo programming are applied in a software project in the academic context. We have three groups. The first group will use solo programming to develop their project. The second group will use pair programming to develop their project. The third group will use a combination of solo and pair programming. We do triangulation to understand how the implementation of hybrid pair programming in global software development affects. Our results show that pair programming cannot be applied in global software development activity. Some activities work well in pairs, and some activities work well solo. In conjunction, As a result, the organization may implement pair programming and solo programming in a global software development environment through our proposed practices, namely hybrid pair programming. These basic findings are consistent with research showing that hybrid pair programming can work well in a global software development environment. However, pair programming provides better productivity than solo programming (55% better). Hybrid pair programming can improve coordination, scheduling, and technical issues and boost solo programming productivity.
KEYWORDS - Global Software Development, Pair Programming, Distributed Software Development
Developing Cross-Platform Library using Multi-OS Engine
Dilkhaz Yaseen Mohammed, Peter Cooper
Libraries are of great importance in the development of mobile apps. Mobile application development services have reached a higher level with APIs. When developers develop applications for mobile devices, they often rely on APIs for connectivity. In fact, APIs accelerate mobile development and provide exceptional agility for organizations undergoing their own digital transformation. As a result of developing a library and then sharing it with the world, others can benefit from it in their own projects. The programmer needs to make sure that he uses APIs available for both the Android and iOS platforms. The programmer\\\\\\\'s creation of an interface for accessing platform-specific functions from the library and creating Android and iOS applications in its projects accelerates the development of software projects. In this paper, the programmer uses Intel’s Multi-OS Engine Technology to give the possibility to use Java capabilities to develop native mobile applications for Apple iOS and Android devices, providing the native look, feel, and performance. This technology provides a stand-alone plug-in that integrates into Android Studio on Windows and Apple Mac OS development machines. When developing a Java open source project, the common conclusion the programmer always ends up with is to share the produced outcomes with the developer community, which should be the least objective in the Java world then utilize the library. Then utilize the library in both Android and iOS apps.
KEYWORDS - Cross-Platform, Java, Retrofit
Fuzzy Expert System: an Intelligence Framework for Diagnosing Malaria
Muzammil Adamu, Muhammad Lawal Jibril
Malaria is a deadly disease killing millions of people every year. Different countries of the world, governmental and non-governmental organizations including World Health Organization have taken it as a challenge to address the issue of deaths associated with malaria. Prompt and accurate diagnosis is a major key in medical field. This prompts for the need to develop a diagnosis and therapy system that will help in diagnosing of malaria. The fuzzy logic was used, and low, average, high, very high are the linguistic variable used. The system interacts with user using plain English based on some arranged rules which are a typical collection of if/then rules. The system was developed using VB.Net and MySQL database. It is believed that this design can help to reduce the congestion we often see in our hospitals by providing solution for sick patients, irrespective of their locations. It would be of great necessity to provide a computerized system that will provide a complementary medical service, such as medical disease diagnosis in places where accessibility is a problem as well as health care facilities where qualified experts are lacking, hence this topic, Fuzzy Expert System an Intelligent Framework for Diagnosing Malaria.
KEYWORDS - Diagnose, Expert, Framework