Abstract The development and use of artificial intelligence (AI) tools have raised several concerns about the effectiveness of assessment practice. Authentic assessment has emerged as one of the key strategies to cope with the rapidly evolving technology. Nevertheless, prior studies argued that authentic assessment as a standalone instrument is insufficient to control the impact of AI. Thus, several studies called for the integration of AI in assessment design. However, it is unclear how academics can deploy the use of AI in authentic assessment. This paper aims to present AI‐enabled authentic assessment strategies through evidence synthesis. Following the PRISMA guidelines, we collected 103 documents from multiple databases and utilised text mining techniques, including co‐occurrence analysis and topic modelling, to synthesise the documents. The results showed that academic integrity, critical thinking, framework, integration of AI, and teaching and learning methods are the dominant themes across the analysis. Most importantly, our findings identified four pathways in which AI can be deployed. Therefore, we propose the DAD (Deployment options, AI role, and Deliverables) conceptual framework to organise knowledge and guide the future design of AI‐enabled authentic assessment. Context and implications Rationale for this study: The emergence and increasing usage of AI technologies have challenged assessment practices in higher education. While a growing number of studies have proposed various solutions, including frameworks, they address a certain discipline or isolated element, such as authenticity or reflective practice. Why the new findings matter: Our results provide a more holistic and theoretically grounded assessment strategy addressing the alignment between AI integration, pedagogical design, and cognitive engagement to evidence learning in an assessment of learning context. Implications for practitioners, policymakers, and researchers: For authenticity, our result emphasises the need for academics, education practitioners, and policymakers to ensure assessment of learning policies is not against the use of AI. As AI continues to evolve, our result suggests that education practitioners need to ensure that assessment design allows the use of AI as both an agent and a tool to mirror ‘real‐world’ professional context without undermining cognitive engagement.

Topic modelling (TM) is a significant natural language processing (NLP) task and is becoming more popular, especially, in the context of literature synthesis and analysis. Despite the growing volume of studies on the use of and versatility of TM, the knowledge of TM development, especially from the perspective of bibliometrics analysis is limited. To this end, this study evaluated TM research using two techniques namely, bibliometrics analysis and TM itself to provide the current status and the pathway for future studies in the TM field. For this purpose, this study used 16,941 documents collected from Scopus database from 2004 to 2023. Results indicate that the publications on TM have increased over the years, however, the citation impact has declined. Furthermore, the scientific production on TM is concentrated in two countries namely, China and the USA. Our findings showed there are several applications of TM that are understudied, for example, TM for image segmentation and classification. This paper highlighted the future research directions, most importantly, calls for increased multidisciplinary research approaches to fully deploy TM algorithms optimally and thus, increase usage in non-computer science subject areas.

Text clustering holds significant value across various domains due to its ability to identify patterns and group related information. Current approaches which rely heavily on a computed similarity measure between documents are often limited in accuracy and interpretability. We present a novel approach to the problem based on a set of evolved search queries. Clusters are formed as the set of documents matched by a single search query in the set of queries. The queries are optimized to maximize the number of documents returned and to minimize the overlap between clusters (documents returned by more than one query). Where queries contain more than one word they are interpreted disjunctively. We have found it useful to assign one word to be the root and constrain the query construction such that the set of documents returned by any additional query words intersect with the set returned by the root word. Not all documents in a collection are returned by any of the search queries in a set, so once the search query evolution is completed a second stage is performed whereby a KNN algorithm is applied to assign all unassigned documents to their nearest cluster. We describe the method and present results using 8 text datasets comparing effectiveness with well-known existing algorithms. We note that as well as achieving the highest accuracy on these datasets the search query format provides the qualitative benefits of being interpretable and modifiable whilst providing a causal explanation of cluster construction.

Macroeconomic factors have a critical impact on banking credit risk, which cannot be directly controlled by banks, and therefore, there is a need for an early credit risk warning system based on the macroeconomy. By comparing different predictive models (traditional statistical and machine learning algorithms), this study aims to examine the macroeconomic determinants’ impact on the UK banking credit risk and assess the most accurate credit risk estimate using predictive analytics. This study found that the variance-based multi-split decision tree algorithm is the most precise predictive model with interpretable, reliable, and robust results. Our model performance achieved 95% accuracy and evidenced that unemployment and inflation rate are significant credit risk predictors in the UK banking context. Our findings provided valuable insights such as a positive association between credit risk and inflation, the unemployment rate, and national savings, as well as a negative relationship between credit risk and national debt, total trade deficit, and national income. In addition, we empirically showed the relationship between national savings and non-performing loans, thus proving the “paradox of thrift”. These findings benefit the credit risk management team in monitoring the macroeconomic factors’ thresholds and implementing critical reforms to mitigate credit risk.

This study explores the comparative performance of cutting-edge AI models, i.e., Finaance Bidirectional Encoder representations from Transsformers (FinBERT), Generatice Pre-trained Transformer GPT-4, and Logistic Regression, for sentiment analysis and stock index prediction using financial news and the NGX All-Share Index data label. By leveraging advanced natural language processing models like GPT-4 and FinBERT, alongside a traditional machine learning model, Logistic Regression, we aim to classify market sentiment, generate sentiment scores, and predict market price movements. This research highlights global AI advancements in stock markets, showcasing how state-of-the-art language models can contribute to understanding complex financial data. The models were assessed using metrics such as accuracy, precision, recall, F1 score, and ROC AUC. Results indicate that Logistic Regression outperformed the more computationally intensive FinBERT and predefined approach of versatile GPT-4, with an accuracy of 81.83% and a ROC AUC of 89.76%. The GPT-4 predefined approach exhibited a lower accuracy of 54.19% but demonstrated strong potential in handling complex data. FinBERT, while offering more sophisticated analysis, was resource-demanding and yielded a moderate performance. Hyperparameter optimization using Optuna and cross-validation techniques ensured the robustness of the models. This study highlights the strengths and limitations of the practical applications of AI approaches in stock market prediction and presents Logistic Regression as the most efficient model for this task, with FinBERT and GPT-4 representing emerging tools with potential for future exploration and innovation in AI-driven financial analytics.

The dominance of social media has added to the channels of bullying for perpetrators. Unfortunately, cyberbullying (CB) is the most prevalent phenomenon in todays cyber world, and is a severe threat to the mental and physical health of citizens. This opens the need to develop a robust system to prevent bullying content from online forums, blogs, and social media platforms to manage the impact in our society. Several machine learning (ML) algorithms have been proposed for this purpose. However, their performances are not consistent due to high class imbalance and generalisation issues. In recent years, large language models (LLMs) like BERT and RoBERTa have achieved state-of-the-art (SOTA) results in several natural language processing (NLP) tasks. Unfortunately, the LLMs have not been applied extensively for CB detection. In our paper, we explored the use of these models for cyberbullying (CB) detection. We have prepared a new dataset (D2) from existing studies (Formspring and Twitter). Our experimental results for dataset D1 and D2 showed that RoBERTa outperformed other models.

The World Health Organisation (WHO) revealed approximately 280 million people in the world suffer from depression. Yet, existing studies on early-stage depression detection using machine learning (ML) techniques are limited. Prior studies have applied a single stand-alone algorithm, which is unable to deal with data complexities, prone to overfitting, and limited in generalization. To this end, our paper examined the performance of several ML algorithms for early-stage depression detection using two benchmark social media datasets (D1 and D2). More specifically, we incorporated sentiment indicators to improve our model performance. Our experimental results showed that sentence bidirectional encoder representations from transformers (SBERT) numerical vectors fitted into the stacking ensemble model achieved comparable F1 scores of 69% in the dataset (D1) and 76% in the dataset (D2). Our findings suggest that utilizing sentiment indicators as an additional feature for depression detection yields an improved model performance, and thus, we recommend the development of a depressive term corpus for future work.

Sentiment analysis is a technique used to understand the public’s opinion towards an event, product, or organization. For example, sentiment analysis can be used to understand positive or negative opinions or attitudes towards electric vehicle (EV) brands. This provides companies with valuable insight into the public’s opinion of their products and brands. In the field of natural language processing (NLP), transformer models have shown great performance compared to traditional machine learning algorithms. However, these models have not been explored extensively in the EV domain. EV companies are becoming significant competitors in the automotive industry and are projected to cover up to 30% of the United States light vehicle market by 2030 In this study, we present a comparative study of large language models (LLMs) including bidirectional encoder representations from transformers (BERT), robustly optimised BERT (RoBERTa), and a generalised autoregressive pre-training method (XLNet) using Lucid Motors and Tesla Motors YouTube datasets. Results evidenced that LLMs like BERT and her variants are off-the-shelf algorithms for sentiment analysis, specifically when fine-tuned. Furthermore, our findings present the need for domain adaptation whilst utilizing LLMs. Finally, the experimental results showed that RoBERTa achieved consistent performance across the EV datasets with an F1 score of at least 92%.

Predicting credit default risk is important to financial institutions, as accurately predicting the likelihood of a borrower defaulting on their loans will help to reduce financial losses, thereby maintaining profitability and stability. Although machine learning models have been used in assessing large applications with complex attributes for these predictions, there is still a need to identify the most effective techniques for the model development process, including the technique to address the issue of data imbalance. In this research, we conducted a comparative analysis of random forest, decision tree, SVMs (Support Vector Machines), XGBoost (Extreme Gradient Boosting), ADABoost (Adaptive Boosting) and the multi-layered perceptron, to predict credit defaults using loan data from LendingClub. Additionally, XGBoost was used as a framework for testing and evaluating various techniques. Moreover, we applied this XGBoost framework to handle the issue of class imbalance observed, by testing various resampling methods such as Random Over-Sampling (ROS), the Synthetic Minority Over-Sampling Technique (SMOTE), Adaptive Synthetic Sampling (ADASYN), Random Under-Sampling (RUS), and hybrid approaches like the SMOTE with Tomek Links and the SMOTE with Edited Nearest Neighbours (SMOTE + ENNs). The results showed that balanced datasets significantly outperformed the imbalanced dataset, with the SMOTE + ENNs delivering the best overall performance, achieving an accuracy of 90.49%, a precision of 94.61% and a recall of 92.02%. Furthermore, ensemble methods such as voting and stacking were employed to enhance performance further. Our proposed model achieved an accuracy of 93.7%, a precision of 95.6% and a recall of 95.5%, which shows the potential of ensemble methods in improving credit default predictions and can provide lending platforms with the tool to reduce default rates and financial losses. In conclusion, the findings from this study have broader implications for financial institutions, offering a robust approach to risk assessment beyond the LendingClub dataset.

An accurate prediction of house prices is a fundamental requirement for various sectors including real estate and mortgage lending. It is widely recognized that a property value is not solely determined by its physical attributes but is significantly influenced by its surrounding neighbourhood. Meeting the diverse housing needs of individuals while balancing budget constraints is a primary concern for real estate developers. To this end, we addressed the house price prediction problem as a regression task and thus employed various machine learning techniques capable of expressing the significance of independent variables. We made use of the housing dataset of Ames City in Iowa, USA to compare support vector regressor, random forest regressor, XGBoost, multilayer perceptron and multiple linear regression algorithms for house price prediction. Afterwards, we identified the key factors that influence housing costs. Our results show that XGBoost is the best performing model for house price prediction.

The higher education (HE) sector benefits every nation's economy and society at large. However, their contributions are challenged by advanced technologies like generative artificial intelligence (GenAI) tools. In this paper, we provide a comprehensive assessment of GenAI tools towards assessment and pedagogic practice and, subsequently, discuss the potential impacts. This study experimented using three assessment instruments from data science, data analytics, and construction management disciplines. Our findings are two-fold: first, the findings revealed that GenAI tools exhibit subject knowledge, problem-solving, analytical, critical thinking, and presentation skills and thus can limit learning when used unethically. Secondly, the design of the assessment of certain disciplines revealed the limitations of the GenAI tools. Based on our findings, we made recommendations on how AI tools can be utilised for teaching and learning in HE.

Recently, peoples awareness of online purchases has significantly risen. This has given rise to online retail platforms and the need for a better understanding of customer purchasing behaviour. Retail companies are pressed with the need to deal with a high volume of customer purchases, which requires sophisticated approaches to perform more accurate and efficient customer segmentation. Customer segmentation is a marketing analytical tool that aids customer-centric service and thus enhances profitability. In this paper, we aim to develop a customer segmentation model to improve decision-making processes in the retail market industry. To achieve this, we employed a UK-based online retail dataset obtained from the UCI machine learning repository. The retail dataset consists of 541,909 customer records and eight features. Our study adopted the RFM (recency, frequency, and monetary) framework to quantify customer values. Thereafter, we compared several state-of-the-art (SOTA) clustering algorithms, namely, K-means clustering, the Gaussian mixture model (GMM), density-based spatial clustering of applications with noise (DBSCAN), agglomerative clustering, and balanced iterative reducing and clustering using hierarchies (BIRCH). The results showed the GMM outperformed other approaches, with a Silhouette Score of 0.80.

The use of generative artificial intelligence (GenAI) in academia is a subjective and hotly debated topic. Currently, there are no agreed guidelines towards the usage of GenAI systems in higher education (HE) and, thus, it is still unclear how to make effective use of the technology for teaching and learning practice. This paper provides an overview of the current state of research on GenAI for teaching and learning in HE. To this end, this study conducted a systematic review of relevant studies indexed by Scopus, using the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines. The search criteria revealed a total of 625 research papers, of which 355 met the final inclusion criteria. The findings from the review showed the current state and the future trends in documents, citations, document sources/authors, keywords, and co-authorship. The research gaps identified suggest that while some authors have looked at understanding the detection of AI-generated text, it may be beneficial to understand how GenAI can be incorporated into supporting the educational curriculum for assessments, teaching, and learning delivery. Furthermore, there is a need for additional interdisciplinary, multidimensional studies in HE through collaboration. This will strengthen the awareness and understanding of students, tutors, and other stakeholders, which will be instrumental in formulating guidelines, frameworks, and policies for GenAI usage.

Sentiment analysis (SA) has received huge attention to understand customer perception, especially in the movie review (IMDB) domain. This is due to the availability of large, labelled dataset. This has enhanced the use and development of machine learning (ML) algorithms ranging from the traditional machine learning algorithms, deep learning algorithms to large language models. The ML models have shown great performances. However, the application of ML methods for SA is limited in service industry like banking, due to the unavailability of large training dataset. Thus, we consider the use of lexicon-based sentiment analysis appropriate. We employ 346,000 Nigeria bank customers’ tweets to develop our corpus and thus, propose SentiLeye, a novel lexicon-based algorithm for sentiment analysis. Our algorithm incorporates corpus-based approach and external lexical resources for sentiment lexicon generation of Pidgin English language terms (a non-English under resourced language). Moreover, we demonstrate the use of verbs and adverbs that express opinion on service experience to improve the performance of lexicon-based sentiment analysis. Results show that SentiLeye outperforms popular off-the-shelf sentiment lexicons with macro F1-score of 76

Stock price prediction is challenging due to global economic instability, high volatility, and the complexity of financial markets. Hence, this study compared several machine learning algorithms for stock market prediction and further examined the influence of a sentiment analysis indicator on the prediction of stock prices. Our results were two-fold. Firstly, we used a lexicon-based sentiment analysis approach to identify sentiment features, thus evidencing the correlation between the sentiment indicator and stock price movement. Secondly, we proposed the use of GRUvader, an optimal gated recurrent unit network, for stock market prediction. Our findings suggest that stand-alone models struggled compared with AI-enhanced models. Thus, our paper makes further recommendations on latter systems.

Stroke is a major cause of death worldwide, resulting from a blockage in the flow of blood to different parts of the brain. Many studies have proposed a stroke disease prediction model using medical features applied to deep learning (DL) algorithms to reduce its occurrence. However, these studies pay less attention to the predictors (both demographic and behavioural). Our study considers interpretability, robustness, and generalisation as key themes for deploying algorithms in the medical domain. Based on this background, we propose the use of random forest for stroke incidence prediction. Results from our experiment showed that random forest (RF) outperformed decision tree (DT) and logistic regression (LR) with a macro F1 score of 94%. Our findings indicated age and body mass index (BMI) as the most significant predictors of stroke disease incidence.

Topic modelling is a prominent task for automatic topic extraction in many applications such as sentiment analysis and recommendation systems. The approach is vital for service industries to monitor their customer discussions. The use of traditional approaches such as Latent Dirichlet Allocation (LDA) for topic discovery has shown great performances, however, they are not consistent in their results as these approaches suffer from data sparseness and inability to model the word order in a document. Thus, this study presents the use of Kernel Principal Component Analysis (KernelPCA) and K-means Clustering in the BERTopic architecture. We have prepared a new dataset using tweets from customers of Nigerian banks and we use this to compare the topic modelling approaches. Our findings showed KernelPCA and K-means in the BERTopic architecture-produced coherent topics with a coherence score of 0.8463.