- ItemApplication of Machine Learning Classification to Detect Fraudulent E‑wallet Deposit Notification SMSes(LINK Centre, University of the Witwatersrand (Wits), Johannesburg, 2020-06-30) Enkono, Fillemon S.; Suresh, NalinaFraudulent e-wallet deposit notification SMSes designed to steal money and goods from m-banking users have become pervasive in Namibia. Motivated by an observed lack of mobile applications to protect users from such deceptions, this study evaluated the ability of machine learning to detect the fraudulent e-wallet deposit notification SMSes. The naïve Bayes (NB) and support vector machine (SVM) classifiers were trained to classify both ham (desired) SMSes and scam (fraudulent) e-wallet deposit notification SMSes. The performances of the two classifier models were then evaluated. The results revealed that the SVM classifier model could detect the fraudulent SMSes more efficiently than the NB classifier.
- ItemA Supplementary Tool for Web-archiving Using Blockchain Technology(LINK Centre, University of the Witwatersrand (Wits), Johannesburg, 2020-06-30) De Villiers, John E.; Calitz, André P.The usefulness of a uniform resource locator (URL) on the World Wide Web is reliant on the resource being hosted at the same URL in perpetuity. When URLs are altered or removed, this results in the resource, such as an image or document, being inaccessible. While web-archiving projects seek to prevent such a loss of online resources, providing complete backups of the web remains a formidable challenge. This article outlines the initial development and testing of a decentralised application (DApp), provisionally named Repudiation Chain, as a potential tool to help address these challenges presented by shifting URLs and uncertain web-archiving. Repudiation Chain seeks to make use of a blockchain smart contract mechanism in order to allow individual users to contribute to web-archiving. Repudiation Chain aims to offer unalterable assurance that a specific file and its URL existed at a given point in time—by generating a compact, non-reversible representation of the file at the time of its non-repudiation. If widely adopted, such a tool could contribute to decentralisation and democratisation of web-archiving
- ItemPilot Testing of an Information Extraction (IE) Prototype for Legal Research(LINK Centre, University of the Witwatersrand (Wits), Johannesburg, 2020-06-30) Scholtz, Brenda; Padayachy, Thashen; Adewoyin, OluwandeThis article presents findings from pilot testing of elements of an information extraction (IE) prototype designed to assist legal researchers in engaging with case law databases. The prototype that was piloted seeks to extract, from legal case documents, relevant and accurate information on cases referred to (CRTs) in the source cases. Testing of CRT extraction from 50 source cases resulted in only 38% (n = 19) of the extractions providing an accurate number of CRTs. In respect of the prototype’s extraction of CRT attributes (case title, date, journal, and action), none of the 50 extractions produced fully accurate attribute information. The article outlines the prototype, the pilot testing process, and the test findings, and then concludes with a discussion of where the prototype needs to be improved.