Recently, big data streams have become ubiquitous due to the fact that a number of applications generate a huge amount of data at a great velocity. This made it difficult for existing data mining tools, technologies, methods, and techniques to be applied directly on big data streams due to the inherent dynamic characteristics of big data. In this paper, a systematic review of big data streams analysis which employed a rigorous and methodical approach to look at the trends of big data stream tools and technologies as well as methods and techniques employed in analysing big data streams. It provides a global view of big data stream tools and technologies and its comparisons. Three major databases, Scopus, ScienceDirect and EBSCO, which indexes journals and conferences that are promoted by entities such as IEEE, ACM, SpringerLink, and Elsevier were explored as data sources. Out of the initial 2295 papers that resulted from the first search string, 47 papers were found to be relevant to our research questions after implementing the inclusion and exclusion criteria. The study found that scalability, privacy and load balancing issues as well as empirical analysis of big data streams and technologies are still open for further research efforts. We also found that although, significant research efforts have been directed to real-time analysis of big data stream not much attention has been given to the preprocessing stage of big data streams. Only a few big data streaming tools and technologies can do all of the batch, streaming, and iterative jobs; there seems to be no big data tool and technology that offers all the key features required for now and standard benchmark dataset for big data streaming analytics has not been widely adopted. In conclusion, it was recommended that research efforts should be geared towards developing scalable frameworks and algorithms that will accommodate data stream computing mode, effective resource allocation strategy and parallelization issues to cope with the ever-growing size and complexity of data.

The informal sector is an integral part of several sub-Saharan African (SSA) countries and plays a key role in the economic growth of these countries. This article used a comparative systematic review to explore the factors that act as drivers to informality in South Africa (SA) and Nigeria, the challenges that impede the growth dynamics of the informal sector, the dominant subsectors, and policy initiatives targeting informal sector providers. A systematic search of Google Scholar, Scopus, ResearchGate was performed together with secondary data collated from grey literature. Using Boolean string search protocols facilitated the elucidation of research questions (RQs) raised in this study. An inclusion and exclusion criteria became necessary for rigour, comprehensiveness and limitation of publication bias. The data collated from thirty-one (31) primary studies (17 for SA and 14 for Nigeria) revealed that unemployment, income disparity among citizens, excessive tax burdens, excessive bureaucratic hurdles from government, inflationary tendencies, poor corruption control, GDP per capita, and lack of social protection survival tendencies all act as drivers to the informal sector in SA and Nigeria. Several challenges are given for both economies and policy incentives that might help sustain and improve the informal sector in these two countries.

Until recently, traditional machine learning techniques (TMLTs) such as multilayer perceptrons (MLPs) and support vector machines (SVMs) have been used successfully for churn prediction, but with significant efforts expended on the configuration of the training parameters. The selection of the right training parameters for supervised learning is almost always experimentally determined in an ad hoc manner. Deep neural networks (DNNs) have shown significant predictive strength over TMLTs when used for churn predictions. However, the more complex architecture of DNNs and their capacity to process huge amounts of non-linear input data demand more time and effort to configure the training hyperparameters for DNNs during churn modeling. This makes the process more challenging for inexperienced machine learning practitioners and researchers. So far, limited research has been done to establish the effects of different hyperparameters on the performance of DNNs during churn prediction. There is a lack of empirically derived heuristic knowledge to guide the selection of hyperparameters when DNNs are used for churn modeling. This paper presents an experimental analysis of the effects of different hyperparameters when DNNs are used for churn prediction in the banking sector. The results from three experiments revealed that the deep neural network (DNN) model performed better than the MLP when a rectifier function was used for activation in the hidden layers and a sigmoid function was used in the output layer. The performance of the DNN was better when the batch size was smaller than the size of the test set data, while the RemsProp training algorithm had better accuracy when compared with the stochastic gradient descent (SGD), Adam, AdaGrad, Adadelta, and AdaMax algorithms. The study provides heuristic knowledge that could guide researchers and practitioners in machine learning-based churn prediction from the tabular data for customer relationship management in the banking sector when DNNs are used.

E-voting is one of the valid use cases of blockchain technology with many blockchain e-voting systems already proposed. But efforts that focus on critical analysis of blockchain e-voting architectures for national elections from stakeholders’ perspectives are mostly lacking in the literature. Therefore, government decision-makers and election stakeholders do not yet have a sufficient basis to understand the potential risks, challenges, and prospects that are associated with blockchain e-voting. This paper demonstrates how the use of the Architecture Trade-off Analysis Method (ATAM) can enable stakeholders in national elections to understand the risks, prospects, and challenges that could be associated with a blockchain e-voting system for national elections. By using a study context of South Africa, a proposed blockchain e-voting architecture was used as a basis to aid election stakeholders to reason on the concept of blockchain e-voting to get them to understand the potential risks, security threats, critical requirements attributes, and weaknesses that could be associated with using blockchain e-voting for national elections. The study found that blockchain e-voting can prevent many security attacks, internal vote manipulation, and promote transparency. However, voter validation and the security of the blockchain architecture are potential weaknesses that will need significant attention.

Organisations must derive adequate business value (BV) from Business Intelligence (BI) adoption to retain their profitability and long-term sustainability. Yet, the nuances that define the realisation of BV from BI are still not understood by many organisations that have adopted BI. This paper aims to foster a deeper understanding of the relationship between Business Intelligence (BI) and business value (BV) by focusing on the theories that have been used, the critical factors of BV derivation, the inhibitors of BV, and the different forms of BV. To do this, a systematic literature review (SLR) methodology was adopted. Articles were retrieved from three scholarly databases, namely Google Scholar, Scopus, and Science Direct, based on relevant search strings. Inclusion and exclusion criteria were applied to select ninety-three (93) papers as the primary studies. We found that the most used theoretical frameworks in studies on BI and BV are the Resource-Based View (RBV), Dynamic Capabilities Theory (DCT), Technology-Organisation-Environment (TOE), and Contingency Theory (CON). The most acknowledged critical factors of BV are skilled human capital, BI Infrastructure, data quality, BI application and usage/data culture, BI alignment with organisational goals, and top management support. The most acclaimed inhibitors of BV are data quality and handling, data security and protection, lack of BI Infrastructure, and lack of skilled human resource capital, while customer intelligence is the most acknowledged form of BV. So far, many theories that are relevant to BI and BV, critical factors, inhibitors, and forms of BV were marginally mentioned in the literature, requiring more investigations. The study reveals opportunities for future research that can be explored to gain a deeper understanding of the issues of BV derivation from BI. It also offers useful insights for adopters of BI, BI researchers, and BI practitioners.

Academic advising of students is an expert task that requires a lot of time, and intellectual investments from the human agent saddled with such a responsibility. In addition, good quality academic advising is subject to availability of experienced and committed personnel to undertake the task. However, there are instances when there is paucity of capable human adviser, or where qualified persons are not readily available because of other pressing commitments, which will make system-based decision support desirable and useful. In this work, we present the design and implementation of an intelligent Course Advisory Expert System (CAES) that uses a combination of rule based reasoning (RBR) and case based reasoning (CBR) to recommend courses that a student should register in a specific semester, by making recommendation based on the student’s academic history. The evaluation of CAES yielded satisfactory performance in terms of credibility of its recommendations and usability.