Journal of Information & Knowledge Management

Polynomial functions $F$ of degree $m$ have a form in the Bernstein basis defined over $l$-dimensional simplex $W$. The Bernstein coefficients exhibit a number of special properties. The function $F$ can be optimised by the smallest and largest Bernstein coefficients (enclosure bounds) over $W$. By a proper choice of barycentric subdivision steps of $W$, we prove the inclusion property of Bernstein enclosure bounds. To this end, we provide an algorithm that computes the Bernstein coefficients over subsimplices. These coefficients are collected in an $l$-dimensional array in the field of computer-aided geometric design. Such a construct is typically classified as a patch. We show that the Bernstein coefficients of $F$ over the faces of a simplex coincide with the coefficients contained in the patch.

This paper utilises data dimension reduction to settle a heavily debated question concerning the dimensionality of political ideology in the Arab World. It relies on recent data available through the World Values Survey to generate a stable solution for the number of important and exciting dimensions defining ordinary citizens’ political attitude structures. The findings of the analysis suggest that in four Arab states, political ideology is multi-dimensional on the mass level. This negates the widespread assumption made about Arab politics where Islam and secularism constitute the only dimension organising voters’ attitudes and behaviours. This is important because many analyses of Middle Eastern politics start with this assumption without questioning its validity. Further, models of political ideology are to be modified when transferred to studying Middle Eastern political attitudes. The single-dimension hypothesis applicable in some Western settings is not attainable in the Arab World.

This paper presents a new image re-ranking approach that can implicitly improve the retrieved images based on the file’s contents and some user-specific actions. In more detail, multiple descriptors are used to describe image files accurately and they do not require user intervention or tuned parameters. Furthermore, each of these descriptors has a weight, which affects the file rank. Unlike existing approaches, descriptor weight is assigned dynamically and changes from one file to another based on the percentage of differences found by the descriptor. Hence, the developed weight mechanism improves the chance of getting the required files significantly. The performance of the developed approach is investigated through several experiments and it has been observed that the approach has the ability of showing the most relevant files at the top of the query results and increases the percentage of the retrieved relevant files.

In this paper, we present a forecasting scheme for the growth of molecular structures from NMR and X-ray Crystallography experimental techniques released every year by employing an autoregressive (AR) process. The proposed scheme maximises the forecasting accuracy by utilising the optimal AR process order. The optimal model order was derived as the model with the least prediction error. Therefore, the proposed scheme has been efficiently employed to model and predict the annual growth of structures-based NMR and X-ray Crystallography experimental data for the next decade 2019–2028 using the time series of the past 43 years of both experimental datasets. The experimental results showed that the optimal model order to estimate both datasets was $AR(2)$ which belongs to a forecasting accuracy of $98\%$, for both datasets. Indeed, such a high level of accuracy referred to the amount of linearity between the consecutive elements of the original times series. Hence, the forecasting results reveals of an exponential increasing behaviour in the future growth in the annual structures released from both NMR and X-ray Crystallography experiments.

A descriptive statistics approach is applied to analyse a newly collected dataset from Young Syrian Refugees in Lebanon. The aim of the paper is to empirically document the uses and gratifications of YSRs smartphone and social media use. Results indicated that the majority of YSRs use Facebook and WhatsApp for communications and entertainment purposes. Frequency analysis was used to determine the rate at which YSRs own and utilise smartphones and Social Networking Sites. Findings indicate that males head of households purchased smartphones for wives, younger sisters and mothers. Cross tabulation analysis of the role of gender on uses and gratifications of smartphones yielded an exclusion effect due to cultural consideration. This paper opens up new horizons in the use of information technology for improving the living conditions of refugee populations.

Data Envelopment Analysis (DEA) has been applied creatively in various study domains to compare and evaluate different Decision Making Units (DMUs) based on multiple input–output attributes. In this paper, the performance of Jordanian public hospitals is assessed via a methodology combining DEA with data mining methods, specifically, clustering. Initially, inputs of inefficient hospitals were altered to check for waste in the allocated resources. Then, the number of inputs–outputs was manipulated to test if the number is strongly influencing the productivity of the DMUs. The number of DMUs used was 27 public hospitals and the applicable efficiency measurements used were constant return to scale (CRS) and variable return to scale (VRS) through the DEAP software. Experiments showed that the efficiency of a hospital might be more meaningfully assessed if it is compared with a group of hospitals that are similar in some factors. More specifically, results of applying the CRS model proved that 77% of the hospitals were efficient. Additionally, we found that the inefficiencies of some hospitals are linked to weak resource utilization. It is concluded that number of inputs–outputs inserted in the efficiency evaluation process impacts the resulted values.

The use of social networks such as Facebook, Twitter, and YouTube in the society has become ubiquitous. The advent of communication technologies alongside other unification trends and notions such as media convergence and digital content allow the users of the social network to integrate these networks in their everyday life. There have been several attempts in the literature to investigate and explain the use of social networks such as Facebook and WhatsApp by university students in the Arab region. However, little research has been done on how university students utilise online audiovisual materials in their academic activities in the UAE. This research aims to elucidate the use of YouTube as a learning aid for university students in the UAE. We adopt the technology acceptance model (TAM) as the theoretical framework for this investigation. A quantitative methodology is employed to answer the research question. Primary data consisting of 221 correspondents were analysed, covering patterns of using YouTube as an academic audiovisual learning aid. Statistical techniques including descriptive, correlations, regression tests were used to analyse the data. The study concluded that students use YouTube as a learning tool for their academic studies and enriching their general knowledge; and there is a positive relationship between the use of YouTube videos in academic settings and the students’ overall performance. This study can shed light for teachers, curriculum designers, government entities, and other stakeholders on how to best utilise and integrate the online technology — YouTube — as a learning aid.

Modern learning techniques require technologies to be embedded within the process. The assessment strategy of testing students has been developed through the past few decades and have presented online computer-based examination systems as a replacement for the conventional paper-based testing. However, a major challenge that is still faced by those examination systems is that there is no any guarantee that students sitting adjacently will have different questions or forms, and the questions of exams’ form are not duplicated for any adjacent students within a lab. In fact, this can influence the reliability pertaining to the online examination systems. Consequently, a new algorithm is proposed in this paper for an online examination system that is capable of effectively generating several exam forms with minimal similarity for the constitutional questions, and that guarantees that two adjacent student’s terminals will not have the same forms. Further, the paper provides an explanation for the proposed algorithm and provides a real-life case for the implementation pertaining to the produced algorithm in a conducted experiment. It is found to be proven from the obtained results that the implementation, which is related to the produced form generation algorithm within this research reveals an evidence that is in line with the research assumptions in which an online system’s assessment reliability can be improved based on eliminating adjacent locations of a form similarity.

Preparing an optimal exam timetable in universities is challenging for head of departments, especially for colleges with multiple number of departments, courses, and students. Harmony search algorithm is used by many researchers to solve this problem but none of them could get an optimal solution. In this paper, a new algorithm which is called optimised harmony search algorithm with distributed selections is proposed by optimising the harmony search algorithm and the genetic algorithm. The new algorithm could satisfy hard, soft, and general constraints and generate an optimal exam timetable for a huge number of courses and students. The proposed algorithm is implemented and applied on Jadara University, the algorithm uses an upper triangular matrix to reduce relationships and memory usage, a three-dimensional matrix to ease the exams timetable management a deterministic number generators to eliminate timeslots conflicts, and backtracking algorithm to enhance the population selections. Lecturers and students’ feedback showed a good satisfaction as well the system results.

The domain of Cloud and Service-oriented requirements engineering (SORE) focuses on modelling, specifying, and analysing requirements to enable software development in a service-oriented manner. Even though Cloud and SORE has received increasing attention over the past decade, there is no clear view of the current state of knowledge. Our goal is to gain insights into the current status of Cloud and SORE research published to date. To this end, we conducted a systematic mapping study to develop a map to classify relevant literature published between 2007 and 2018. Out of the 326 studies that were examined, we selected 24 studies as primary studies. The resulting map provides an overview of the existing work on Cloud and SORE and makes it possible to identify gaps and plan and position future research. Our findings suggest that Cloud and SORE as a research context needs specific attention. More evaluation and validation research is needed to better understand the implementation and impact of the various solutions (e.g. frameworks, tools) that are proposed in the Cloud and SORE context.

This research developed a novel algorithm to evaluate internet-based services such as VoIP, Video Conferencing, HTTP and FTP, of different IEEE 802.11 technologies in order to identify the optimum network architecture among Basic Service Set (BSS), Extended Service Set (ESS) and the Independent Basic Service Set (IBSS). The proposed algorithm will yield the rank order of different IEEE 802.11 technologies. By selecting the optimum network architecture and technology, the best overall network performance that provides good voice, video and data quality is guaranteed. Furthermore, it meets the acceptance threshold values for the VoIP, Video Conferencing, HTTP and FTP quality metrics. This algorithm was applied to various room sizes ranging from $2\times 3$m to $10\times 14$m and the number of nodes ranged from 1 to 65. The spatial distributions considered were circular, uniform and random. The Quality of Service (QoS) metrics used were delay, jitter, throughput and packet loss.

This study aims to investigate the role of social media on consumer behaviour in the restaurants industry of Bahrain. The objectives of this research were to assess importance of social media as a marketing tool for Bahrain’s restaurants for influencing consumer purchase decision-making. Such an objective was reached by the scholars who randomly collected data from 224 respondents in Bahrain, who use social media, using a survey instrument. The findings of the descriptive analysis revealed that social media affects consumers’ decision-making, such that the consumers use the social media to study new products, services, and particularly new brands. Also, ongoing consumer discussions reveal how social media exposes the pros and cons of products, services, and brands, so consumers are comfortable with their online decisions, since they can select broads before settling on what is the best price, quality, etc. Moreover, the social media speeds the spread of information unlike any other mass media; ensuring that new knowledge reaches consumers, hence enhancing competition among retailers in the restaurant industry. This study concludes that social media plays a vital role in promoting consumer welfare and decision-making.

High-dimensional data poses unique challenges in outlier detection process. Most of the existing algorithms fail to properly address the issues stemming from a large number of features. In particular, outlier detection algorithms perform poorly on dataset of small size with a large number of features. In this paper, we propose a novel outlier detection algorithm based on principal component analysis and kernel density estimation. The proposed method is designed to address the challenges of dealing with high-dimensional data by projecting the original data onto a smaller space and using the innate structure of the data to calculate anomaly scores for each data point. Numerical experiments on synthetic and real-life data show that our method performs well on high-dimensional data. In particular, the proposed method outperforms the benchmark methods as measured by $F_1$-score. Our method also produces better-than-average execution times compared with the benchmark methods.

Machine learning (ML) is a branch of computer science that is rapidly gaining popularity within the healthcare arena due to its ability to explore large datasets to discover useful patterns that can be interepreted for decision-making and prediction. ML techniques are used for the analysis of clinical parameters and their combinations for prognosis, therapy planning and support and patient management and wellbeing. In this research, we investigate a crucial problem associated with medical applications such as autism spectrum disorder (ASD) data imbalances in which cases are far more than just controls in the dataset. In autism diagnosis data, the number of possible instances is linked with one class, i.e. the no ASD is larger than the ASD, and this may cause performance issues such as models favouring the majority class and undermining the minority class. This research experimentally measures the impact of class imbalance issue on the performance of different classifiers on real autism datasets when various data imbalance approaches are utilised in the pre-processing phase. We employ oversampling techniques, such as Synthetic Minority Oversampling (SMOTE), and undersampling with different classifiers including Naive Bayes, RIPPER, C4.5 and Random Forest to measure the impact of these on the performance of the models derived in terms of area under curve and other metrics. Results pinpoint that oversampling techniques are superior to undersampling techniques, at least for the toddlers’ autism dataset that we consider, and suggest that further work should look at incorporating sampling techniques with feature selection to generate models that do not overfit the dataset.

Background and objective: Chronic kidney disease (CKD) is one of the deadly diseases that can affect a lot of vital organs in the human body such as heart, liver, and lungs. Many individuals might be at early stage of kidney disease and not have any signs, which might lead to a sudden death. Previous research showed that early prediction of CKD is very important in the medical field for physicians’ decision-making and patients’ health and life. To this end, constructing an efficient prediction system for CKD, which is the goal of this paper, often reduces medical errors and overall healthcare cost. Methods: Classification and association rule mining techniques were integrated and utilised to construct an efficient system for predicting and diagnosing CKD and its causes using weka and SPSS as platform environments. In particular, five classification algorithms, namely, naive Bayes, decision tree, support vector machine, K-nearest neighbour, and JRip were used to achieve the research goal. In addition, Apriori algorithm was used to discover strong relationship rules between attributes. The experiments were conducted on real medical dataset collected from hospitals and patient monitoring systems. Results: The experiments achieved high accuracy of 98.50% for K-nearest neighbour (KNN) classifier and achieved 96.00% when using classier based on association rule (JRip). Conclusions: We conclude by showing that applying integrative approach by combining classification algorithms and association rule mining can significantly improve the classification accuracy and be more useful for CKD prediction. This research has also several theoretical and practical implications for the medical field and healthcare industry.

Data imbalance with respect to the class labels has been recognised as a challenging problem for machine learning techniques as it has a direct impact on the classification model’s performance. In an imbalanced dataset, most of the instances belong to one class, while far fewer instances are associated with the remaining classes. Most of the machine learning algorithms tend to favour the majority class and ignore the minority classes leading to classification models being generated that cannot be generalised. This paper investigates the problem of class imbalance for a medical application related to autism spectrum disorder (ASD) screening to identify the ideal data resampling method that can stabilise classification performance. To achieve the aim, experimental analyses to measure the performance of different oversampling and under-sampling techniques have been conducted on a real imbalanced ASD dataset related to adults. The results produced by multiple classifiers on the considered datasets showed superiority in terms of specificity, sensitivity, and precision, among others, when adopting oversampling techniques in the pre-processing phase.

Multi-label classification (MLC) generalises the conventional binary and multi-class classification by allowing instances to be linked with one or more of the class labels. Therefore, class labels in MLC are not mutual exclusive as in single label classification (SLC). Consequently, the search space of the MLC problem is large compared with that of SLC and grows in an exponential way. One main approach to solve MLC problem is through forcing instances to be associated with only one class label. This approach of handling the problem of MLC has been widely known as problem transformation method (PTM). Existing PTMs depend on the frequency of class labels as a transformation criterion, which causes several problems such as imbalance class distribution, complicating the training phase and most importantly decreasing the accuracy of the classification task. Therefore, in this paper, a new PTM is proposed based on the positive local dependencies among labels. The proposed PTM aims to facilitate capturing the most accurate positive dependencies among labels and hence improve the predictive performance of the classification task. Experiments on several datasets revealed the superiority of the proposed PTM compared with the existing PTMs, especially with high cardinality datasets.

Sentiment classification is the process of classifying emotions and opinions in texts. In this study, the problem of Arabic sentiment analysis was addressed. A support vector machine (SVM) model was proposed to classify opinions in Arabic micro-texts as being positive or negative. To evaluate the performance of the SVM model, a dataset was built from tweets discussing several social issues in Saudi Arabia. These issues include changes that were implemented by the country as part of a newly established vision, known as Saudi Arabia Vision 2030. The constructed dataset was manually annotated according to the sentiment conveyed in the text. To achieve the best sentiment classification accuracy, several procedures were implemented within the proposed framework including light stemming, feature extraction (Ngrams, emoji and tweet-topic features), parameter optimisation and feature-set reduction. The experimental results revealed excellent outcomes. An accuracy of 89.83% was achieved using the proposed SVM model.

Feature selection, the process of identifying relevant features to be incorporated into a proposed model, is one of the significant steps of the learning process. It removes noise from the data to increase the learning performance while reducing the computational complexity. The literature review indicated that most previous studies had focused on improving the overall classifier performance or reducing costs associated with training time during building of the classifiers. However, in this era of big data, there is an urgent need to deal with more complex issues that makes feature selection, especially using filter-based methods, more challenging; this in terms of dimensionality, data structures, data format, domain experts’ availability, data sparsity, and result discrepancies, among others. Filter methods identify the informative features of a given dataset to establish various predictive models using mathematical models. This paper takes a new route in an attempt to pinpoint recent practical challenges associated with filter methods and discusses potential areas of development to yield better performance. Several practical recommendations, based on recent studies, are made to overcome the identified challenges and make the feature selection process simpler and more efficient.

Road-type classification is increasingly becoming important to be embedded in interactive maps to provide additional useful information for users. The ubiquity of smartphones supported with high definition cameras offers a rich source of information that can be utilised by machine learning techniques. In this paper, we propose a novel Convolutional Neural Network (CNN)-based approach to classify road types using a collection of publicly available images. To overcome the challenge of having huge dataset to train and test CNNs, our approach employs fine-tuning. We conducted an experiment where the VGG-16, VGG-S and GoogLeNet networks were constructed and fine-tuned with the dataset gathered. Our approach achieved an accuracy of 99% in VGG-16 and 100% in VGG-S, while using the GoogLeNet model produced results up to 98%.

Patient satisfaction in the developing world has not received sufficient interest in the healthcare literature. More importantly, models predicting patient satisfaction seldom incorporates patient perspectives into their measurement providing ill-suited metrics for healthcare administrators to make decisive decisions concerning service quality improvements. The current analysis utilizes original data collected from 336 patients in Jordan. This paper tests whether service quality (SERVQUAL) is suitable to be used in the new context and concluded that SERVQUAL is reliable and valid in supplying stakeholders with reliable and valid information concerning patient satisfaction. Further, the analysis suggests a large gap between perceived and expected services performed by Jordanian hospitals signalling the need to service quality enhancement. The study calls for the adoption of agile and rapid patient screening models in emergency departments to reduce wait times and decrease the number of patients left without being seen. This study constitutes an emerging attempt at better understanding how service quality influences patient satisfaction, and more importantly, introduce affordable and cost-effective solutions capable of improving satisfaction in the short run.

This study aims at examining the mediating effect of tax management on the relationship between certain board characteristics including independence, gender diversity, and activeness and firm performance. The sample of the study includes 135 companies in the industrial and service industries listed with Amman Stock Exchange in 2008–2017. This study selects purposive samples using the panel data technique. Results indicate that first, tax management mediates the board independence relationship with financial performance. Second, tax management does not mediate the board gender diversity and board activities relationships respectively with financial performance.

This research aims to investigate the effects of social media use on higher education teaching and learning as well as the students’ academic performance. A total of 275 students and faculty members from the College of Computer Science and Information Technology at Imam Abdulrahman Bin Faisal University took part in the study. The participants answered survey questions to analyse information on their use of social media in education and how that has affected their teaching, learning and grades. A majority of the participants reported that they used social media in training. However, they also stated that social media platforms were beneficial in academic matters. The number of participants who stated that the use of social media in learning helped improve their grades was 43%. The other 57% thought that social media had no impact on their grades or had an adverse effect or were undecided.

Recently, organisations are viewed as a network of connected individuals embedded within a larger network. This research aims to introduce a Network Thinking framework that applies the proven efficiencies of computer networking concepts on human networks to address the issues faced by organisations at different levels. Organisational Network Analysis (ONA) methods have been used to analyse the informal network of interactions between members of the system delivery department within a financial institution. The ONA tools were used to simulate the risks faced by the organisation as a result of the current network of interactions between individuals and the impact of Network Thinking approaches on addressing these issues. This research shows that Network Thinking concepts can be applied on human networks but cannot completely control the behaviour of its nodes. The performance is influenced by how nodes interact with each other and the degree of their accessibility to network resources.

This study aims to assess why users are willing/hesitant to continue using FinTech services based on their perceived benefits and risks pertaining to the use of FinTech technology. Data was collected, using an adopted survey instrument, from bankers based in Bahrain, the financial and FinTech hub of the Middle East. Data analysis was applied to assess the reliability and validity of this study’s conceptual model along with its nine hypotheses with 374 valid responses subsequently being analysed using multiple regression via SPSS version 23. The empirical findings of this study supported all the hypotheses, revealing that both perceived benefit and risks affect the intent to continue using FinTech. Perceived benefit has a stronger effect than the perceived risk, and convenience perceived most beneficial while financial risk perceived riskiest for using FinTech technology by the bankers of Bahrain.

As the world grows and develops, people become more aware of business operational processes that are extremely harmful to our environment. Recently, customers’ demands, and governmental legislations have forced domestic and global organizations to implement environmentally sustainable practices in order to cure what has been damaged in the previous decades. This study focuses on the impact of green supply chain management practices on firm’s performance. These practices are green purchasing, green packaging, green manufacturing, green design and green marketing. A questionnaire was circulated among two companies of two industries: shipping and automobile. 80 managers and supervisors were targeted; 59 responses were received indicating a response rate of (74%). For testing, regression analysis and T-test were used. From the analysis, it was found that most of green practices influence firm’s performance with green manufacturing the most adopted practice that shapes the industries, followed by green marketing and green purchasing. On the other hand, little impact was found in green packaging and negative impact of green design on firm’s performance. Overall, the study indicates that the tested sample is moving towards the green concept but still it is in its first stages and needs more efforts and support from firms, suppliers and customers.