Abstract4This paper reports on the evaluation of Deep Learning (DL) transformer architecture models for Named-Entity Recognition (NER) on ten low-resourced South African (SA) languages.In addition, these DL transformer models were compared to other Neural Network and Machine Learning (ML) NER models.The findings show that transformer models substantially improve performance when applying discrete finetuning parameters per language.Furthermore, fine-tuned transformer models outperform other neural network and machine learning models on NER with the low-resourced SA languages.For example, the transformer models obtained the highest F-scores for six of the ten SA languages and the highest average F-score surpassing the Conditional Random Fields ML model.Practical implications include developing high-performance NER capability with less effort and resource costs, potentially improving downstream NLP tasks such as Machine Translation (MT).Therefore, the application of DL transformer architecture models for NLP NER sequence tagging tasks on low-resourced SA languages is viable.Additional research could evaluate the more recent transformer architecture models on other Natural Language Processing tasks and applications, such as Phrase chunking, MT, and Part-of-Speech tagging.

Neural Network (NN) models produce state-ofthe-art results for natural language processing tasks.Further, NN models are used for sequence tagging tasks on lowresourced languages with good results.However, the findings are not consistent for all low-resourced languages, and many of these languages have not been sufficiently evaluated.Therefore, in this paper, transformer NN models are used to evaluate named-entity recognition for ten low-resourced South African languages.Further, these transformer models are compared to other NN models and a Conditional Random Fields (CRF) Machine Learning (ML) model.The findings show that the transformer models have the highest F-scores with more than a 5% performance difference from the other models.However, the CRF ML model has the highest average F-score.The transformer model's greater parallelization allows lowresourced languages to be trained and tested with less effort and resource costs.This makes transformer models viable for low-resourced languages.Future research could improve upon these findings by implementing a linear-complexity recurrent transformer variant.

Agile software development methodologies are used in many industries of the global economy.The Scrum framework is the predominant Agile methodology used to develop, deliver, and maintain complex software products.While the success of software projects has significantly improved while using Agile methodologies in comparison to the Waterfall methodology, a large proportion of projects continue to be challenged or fails.The primary objective of this paper is to use machine learning to develop predictive models for Scrum adoption, identifying a preliminary model with the highest prediction accuracy.The machine learning models were implemented using multiple linear regression statistical techniques.In particular, a full feature set adoption model, a transformed logarithmic adoption model, and a transformed logarithmic with omitted features adoption model were evaluated for prediction accuracy.Future research could improve upon these findings by incorporating additional model evaluation and validation techniques.

Scrum is the most adopted Agile methodology.The research conducted on Scrum adoption is mainly qualitative and there is therefore a need for a quantitative study on Scrum adoption challenges.The primary objective of this paper is to present the findings of a study on the factors that have a significant relationship with Scrum adoption as perceived by Scrum practitioners working within South African organizations.Towards this objective, a narrative review to extract and synthesize the existing challenges was conducted.These synthesized challenges were used in the development of a conceptual framework for evaluating the challenges that have a correlation and linear relationship with Scrum adoption.Following this, a survey questionnaire was used to test and evaluate the factors forming part of the developed framework.The findings indicate that Relative Advantage, Complexity, and Sprint Management are factors that have a significant linear relationship with Scrum adoption.Our recommendation is that organizations consider these findings during their adoption phase of Scrum.

Scrum has been the most widely adopted Agile methodology over the past decade with Scrum and Scrum variants offering alternatives to the old software development methods.While Scrum plays an important role in the success of Agile development, it does come with its own challenges.In previous research challenges have been analyzed at the organizational and team level, primarily via case studies.However, fundamentally, Scrum needs to be adopted at the individual level.Furthermore, challenges such as inexperience, poor communication, specialization, lack of teamwork, low-quality, organizational culture and Scrum compatibility, have been identified as contributors.This paper therefore discusses the Scrum and Agile adoption challenges faced both globally as well as within the South African borders, from the findings of a narrative review.Secondly, a custom model adapted from the Diffusion of Innovation theoretical model was developed to detect the Scrum adoption challenges experienced within software organizations at the individual level.The custom model referred to as the Scrum Adoption Challenges Detection Model (SACDM) consists of four constructs, namely; individual factors, team factors, organizational factors and technology factors.The constructs are composed of nineteen independent variables that assists in understanding which factors contributes towards an individual either adopting or rejecting Scrum within a software organization.SACDM is therefore used to detect the adoption or rejection of Scrum as the dependent variable based on the independent variables being tested within the four constructs.The model can further be used with a survey questionnaire to provide generalized awareness of Scrum adoption challenges allowing software organizations to make more informed decisions when adopting Scrum.Future research is to allow the model to contribute towards Scrum adoption challenges predictive analysis.