Research Article
Artificial Intelligence and the Decline of Source Verification Practices in Higher Education in Western Uganda
Kimulya Adorable*
Issue:
Volume 1, Issue 4, December 2026
Pages:
130-146
Received:
6 July 2026
Accepted:
23 July 2026
Published:
12 November 2026
DOI:
10.11648/j.sdai.20260104.11
Downloads:
Views:
Abstract: The increasing embrace and utilization of artificial intelligence (AI) within higher education has expressively altered the ways academic knowledge is created, retrieved, and shared. AI-powered tools have become common resources for both students and teachers. The tools generate clear explanations, summarized work, and good-looking content. While these tools provide notable, the growing overdependence upon them has greatly eroded scholarly practices of both students and teachers. The key challenge is the lack of AI content verification practices. In many university settings, the accessibility of AI tools through internet-enabled devices has reduced the need to consult libraries or engage with multiple scholarly sources because users find AI-generated responses so swift, with ready-made content that does not overwork them. The study investigated the growing use of artificial intelligence and the decline of knowledge verification practices among students and teachers in higher education institutions in mid-western Uganda. It explores how AI tools are influencing academic habits, including reduced library engagement, increased reliance on instant digital responses, and the growing perception that information obtained from the internet is fundamentally correct. On the contrary, some content may contain incomplete, misleading, or contextually inappropriate information based on the prompts or instructions given to the system, available training data, or algorithmic limitations. Using a mixed-methods approach involving questionnaires and interviews, the study examined the extent to which students and teachers verify AI-generated content before incorporating it into academic work. The findings of this study highlight the emerging culture of unquestionable trust in AI and its implications for scholarly rigor, critical thinking, and academic integrity in Ugandan universities. The study contributes to ongoing debates on artificial intelligence and digital transformation in higher education in Uganda and globally.
Abstract: The increasing embrace and utilization of artificial intelligence (AI) within higher education has expressively altered the ways academic knowledge is created, retrieved, and shared. AI-powered tools have become common resources for both students and teachers. The tools generate clear explanations, summarized work, and good-looking content. While t...
Show More
Research Article
A Systematic Comparison of Deep Learning Architectures for Pipeline Degradation Prediction: LSTM, CNN, and GNN Applied to Nigeria's Oil and Gas Sector
Muhammad Bala Maradun*,
Umar Usman,
Yakubu Musa,
Abdulkarim Bello
Issue:
Volume 1, Issue 4, December 2026
Pages:
147-159
Received:
2 July 2026
Accepted:
25 July 2026
Published:
12 November 2026
DOI:
10.11648/j.sdai.20260104.12
Downloads:
Views:
Abstract: Selecting an appropriate deep learning architecture for pipeline Remaining Useful Life (RUL) prediction requires a systematic comparison under consistent experimental conditions. This revision reports results computed directly on the dataset (500 records spanning 12 installation cohorts from 2000–2024; see Section 3.1). We benchmark Long Short-Term Memory (LSTM) networks, Convolutional Neural Networks (CNN), and Graph Neural Networks (GNN) against a common Artificial Neural Network (ANN) baseline using identical preprocessing and seven-fold stratified cross-validation (Section 3.1 explains why the originally proposed expanding-window temporal cross-validation could not be applied to the data received). The ANN, LSTM, and GNN are statistically distinguishable but practically tied at near-perfect regression performance (GNN: R2 = 0.9998; ANN: R2 = 0.9996; LSTM: R2 = 0.9986; all p < 0.05 pairwise, Section 7.4) and all three achieve perfect binary classification (Accuracy = Precision = Recall = F1 = 1.000). The CNN performs far worse on both tasks (R2 = 0.1655; classification F1 = 0.373), which we attribute to its translation-invariant, weight-shared convolutional structure being poorly suited to a tabular prediction task in which each input feature requires its own, non-interchangeable coefficient (Section 8.2). Direct analysis of the target variable shows that Remaining Useful Life is an exact deterministic linear function of five input covariates (R2 = 1.0000 for ordinary linear regression; Section 3.1.1), which explains why every architecture capable of learning feature-specific weights (ANN, LSTM, GNN) recovers it almost perfectly, and an ablation test shows the GNN’s graph-convolution step contributes negligibly (R2 changes from 0.9998 to 0.9995 when message-passing is disabled), providing no support for the originally hypothesised spatial-relationship-modelling advantage. These findings substantially revise the architecture-selection guidance proposed in the original submission: rather than recommending the GNN as the default architecture, we recommend the ANN as the default choice on the grounds of statistically indistinguishable accuracy at a fraction of the computational cost (Section 3.5), and we recommend against the standard (translation-invariant) CNN for tabular pipeline-attribute prediction tasks of this kind.
Abstract: Selecting an appropriate deep learning architecture for pipeline Remaining Useful Life (RUL) prediction requires a systematic comparison under consistent experimental conditions. This revision reports results computed directly on the dataset (500 records spanning 12 installation cohorts from 2000–2024; see Section 3.1). We benchmark Long Short-Term...
Show More
