Forthcoming Articles

International Journal of Artificial Intelligence and Soft Computing

International Journal of Artificial Intelligence and Soft Computing (IJAISC)

Forthcoming articles have been peer-reviewed and accepted for publication but are pending final changes, are not yet published and may not appear here in their final order of publication until they are assigned to issues. Therefore, the content conforms to our standards but the presentation (e.g. typesetting and proof-reading) is not necessarily up to the Inderscience standard. Additionally, titles, authors, abstracts and keywords may change before publication. Articles will not be published until the final proofs are validated by their authors.

Forthcoming articles must be purchased for the purposes of research, teaching and private study only. These articles can be cited using the expression "in press". For example: Smith, J. (in press). Article Title. Journal Title.

Articles marked with this shopping trolley icon are available for purchase - click on the icon to send an email request to purchase.

Online First articles are also listed here. Online First articles are fully citeable, complete with a DOI. They can be cited, read, and downloaded. Online First articles are published as Open Access (OA) articles to make the latest research available as early as possible.

Open AccessArticles marked with this Open Access icon are Online First articles. They are freely available and openly accessible to all without any restriction except the ones stated in their respective CC licenses.

Register for our alerting service, which notifies you by email when new issues are published online.

International Journal of Artificial Intelligence and Soft Computing (3 papers in press)

Regular Issues

  • BD Currency Detection: A CNN-Based Approach with Mobile App Integration   Order a copy of this article
    by M.D. Zahurul Haque, Syed Jubayer Jaman, Md Robiul Islam, Usama Abdun Noor 
    Abstract: Automated currency recognition systems are critical for enhancing financial security, facilitating commerce, and providing assistive solutions for visually impaired populations. Conventional approaches, including manual verification and optical scanning techniques, exhibit inherent limitations in accuracy, processing speed, and adaptability. This paper presents a robust currency recognition framework leveraging Convolutional Neural Networks (CNNs) for the precise classification of Bangladeshi banknotes. A comprehensive dataset of 50,000 annotated images was curated, preprocessed, and augmented to train a deep CNN architecture optimized for multi-class currency denomination classification. The proposed model demonstrated exceptional performance, achieving a training accuracy of 99.77% and validation accuracy of 100%, substantially outperforming traditional image processing-based methods. To facilitate practical deployment, the trained model was optimized and converted to TensorFlow Lite format, enabling efficient inference on resource-constrained mobile devices. An Android-based mobile application was developed to provide real-time, offline currency recognition capabilities. Experimental results demonstrate the efficacy of deep learning methodologies in addressing currency recognition challenges, offering a scalable, accurate, and accessible solution with significant implications for financial inclusion and assistive technology applications.
    Keywords: Convolutional Neural Networks; Currency Recognition; Bangladeshi Banknotes; Deep Learning; TensorFlow Lite; Mobile Application; Assistive Technology.
    DOI: 10.1504/IJAISC.2026.10081214
     
  • Strategic AI Funding Allocation in Sri Lankas Legal Sector: A Resource-Based View Approach   Order a copy of this article
    by Sithara P. Menikpura, Qi’an LIU, Chiagoziem C. Ukwuoma, Charitha N. Menikpura, Chinedu I. Otuka, Chibueze D. Ukwuoma 
    Abstract: This study examines a strategic framework for integrating Artificial Intelligence (AI) into Sri Lankan industries, with a particular focus on the legal sector, utilising the Resource-Based View (RBV) model Despite AI's potential to revolutionise legal research, case management, and access to justice, its adoption in Sri Lanka remains limited due to infrastructural disparities and resource constraints, particularly in rural areas Drawing on empirical data, including survey results and a VRIO (Valuable, Rare, Inimitable, Organised) analysis, this research develops a financing prioritisation matrix to guide AI investment policies Through regression and cost-benefit analyses, the study identifies high-return legal domains, such as corporate and criminal law, and proposes targeted subsidies and training programs for under-resourced regions The findings advocate for a cascading policy agenda to promote the ethical, inclusive, and efficient adoption of AI, maximising socio-economic benefits while bridging urban-rural divides.
    Keywords: Artificial Intelligence; Resource-Based View; VRIO Framework; Strategic Funding; Sri Lanka; Legal Sector.
    DOI: 10.1504/IJAISC.2026.10081501
     
  • A Fermi-Dirac Similarity Function as a Geometric Split-Quality Scorer for Random Forest Classification: An Indefinite Function Beyond the Reproducing-Kernel-Hilbert-Space Framework   Order a copy of this article
    by Kala Nisha G, Hari Hara Krishna Kumar Viswanathan, Nitish Karthikeyan 
    Abstract: Standard random forest splitting criteria Gini impurity, entropy, and log-loss evaluate candidates using class proportions alone, ignoring the geometric arrangement of observations in feature space. We propose a Fermi- Dirac similarity function from quantum statistical mechanics as a geometric split-quality scorer, tested under two architectures: pure replacement of Gini impurity, and a two-stage design that re-ranks Gini-shortlisted candidates by Fermi-Dirac separation. Eigenvalue analysis confirms the Fermi-Dirac Gram matrix is genuinely indefinite (negative-eigenvalue mass 14.8%34.2%). Despite violating positive semi-definiteness, the two-stage architecture attains the highest G-Mean on two of nine datasets, is statistically indistinguishable from baselines under the Friedman-Nemenyi test, and dominates pure replacement on eleven of twelve benchmarks. Positive semi-definiteness is thus irrelevant when a similarity function operates via arithmetic aggregation. Supplementary experiments show the design excels on geometric benchmarks but degrades on high-dimensional sparse and mixed-type data, positioning it as a conditional alternative, not a replacement
    Keywords: Random forest; split criterion; Fermi-Dirac function; indefinite similarity.
    DOI: 10.1504/IJAISC.2026.10081725
     

 

Return to top