Forthcoming Articles

International Journal of Computational Science and Engineering

International Journal of Computational Science and Engineering (IJCSE)

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 Articles will not be published until the final proofs are validated by their authors.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 Computational Science and Engineering (5 papers in press)

Regular Issues

  • Self-supervised learning with split batch repetition strategy for long-tail recognition   Order a copy of this article
    by Zhangze Liao, Liyan Ma, Xiangfeng Luo, Shaorong Xie 
    Abstract: Deep neural networks cannot be well applied to balance testing when the training data present a long tail distribution. Existing works improve the performance of the model in long tail recognition by changing the model training strategy, data expansion, and model structure optimisation. However, they tend to use supervised approaches when training the model representations, which makes the model difficult to learn the features of the tail classes. In this paper, we use self-supervised representation learning (SSRL) to enhance the model's representations and design a three-branch network to merge SSRL with decoupled learning. Each branch adopts different learning goals to enable the model to learn balanced image features in the long-tail data. In addition, we propose a Split Batch Repetition strategy for long-tailed datasets to improve the model. Our experiments on the Imbalance CIFAR-10, Imbalance CIFAR-100, and ImageNet-LT datasets outperform existing similar methods. The ablation experiments prove that our method performs better on more imbalanced datasets. All experiments demonstrate the effectiveness of incorporating the self-supervised representation learning model and split batch repetition strategy.
    Keywords: long-tail recognition; self-supervised learning; decoupled learning; image classification; deep learning; neural network; computer vision;.

  • INTENTIA: continuous-time semantic drift modelling for predictive intent compliance in SDN   Order a copy of this article
    by Anjana Jimmington, Sabu M. Thampi 
    Abstract: Intent-Based Networking (IBN) allows operators to specify high-level network objectives while delegating enforcement to the control plane. Existing IBN monitoring mechanisms evaluate intent compliance using snapshot-based observations, implicitly assuming that intent semantics remain invariant over time. In dynamic network environments, traffic fluctuations and resource contention progressively alter network behaviour, leading to gradual intent degradation that is not immediately observable. This paper presents INTENTIA, a framework that models intent compliance as a continuous-time latent semantic state to capture the temporal evolution of intent satisfaction. By estimating semantic drift trajectories, the framework predicts potential intent violations prior to explicit failures and provides interpretable early-warning indicators for proactive control actions. INTENTIA is evaluated through dataset-driven analysis and controller-in-the-loop SDN emulation, achieving an AUC of 0.986 for violation prediction at a five-second horizon with inference latency below 35 ms, demonstrating improved early detection capability over snapshot-based monitoring approaches.
    Keywords: intent-based networking; IBN; software-defined networking; SDN; semantic drift; continuous-time modelling; probabilistic forecasting; predictive intent monitoring; network management.
    DOI: 10.1504/IJCSE.2026.10081335
     
  • A decentralised platform leveraging blockchain and quantum-secure communication for scientific diplomacy and trusted resource sharing   Order a copy of this article
    by Bogdan Tiganoaia, Ionut-Petrisor Anghel, Doina Banciu 
    Abstract: This paper proposes a decentralized platform specifically designed to facilitate scientific diplomacy and secure resource sharing by integrating blockchain technology with a simulated distribution of quantum keys. The blockchain module is built on the Ethereum network using Solidity smart contracts which record immutable metadata for each resource uploaded through the INFURA service, thereby guaranteeing data traceability and integrity. Authentication is achieved through J-PAKE mechanism extended with a quantum key distribution (QKD) simulation based on the IBM-Qiskit SDK. In parallel, the IPFS ensures encrypted storage of learning materials within a decentralized network, eliminating single points of vulnerability. Unlike traditional platforms which rely on centralized infrastructures, the proposed solution adopts a hybrid blockchain-quantum approach, reducing the risks of security breaches and unauthorized access. This work provides proof of concept for the integration of quantum protocol-simulation-assisted session key derivation with the immutability of blockchain and the advantages of decentralized storage in a modern context.
    Keywords: blockchain; quantum key distribution; QKD; Flask; IBM Qiskit SDK; Ethereum; IPFS; solidity; Infura; react/vue; scientific diplomacy.
    DOI: 10.1504/IJCSE.2026.10082012
     
  • Multimodal sentiment analysis based on text-guided feature decomposition and dual cross-modal interaction   Order a copy of this article
    by Liang Sun, Jie Li, Guangli Zhu, Yuchao Zhu, Shunxiang Zhang, Yang Xu 
    Abstract: Multimodal sentiment analysis often underutilizes text, causing blurred modal boundaries and semantic noise. This leads to sentiment representations that deviate from true meanings, highlighting the need to better exploit text's dominant role in feature decomposition and cross-modal interactions. To address these limitations, we propose a multimodal sentiment analysis framework based on text-guided feature decomposition and dual cross-modal interaction. We introduce text-guided contrastive learning into feature decomposition to construct a text-guided feature decomposition module. This enhances modal discrimination capabilities, improves modal consistency, and prevents blurred modal boundaries. Further, we introduce a text-guided bidirectional cross-modal attention mechanism into the cross-modal interaction process to construct a dual cross-modal interaction module. This module serves to filter introduced noise and to refine and calibrate sentiment representations that deviate from their true semantic meanings. Experimental results on two public multimodal datasets show that the proposed method outperforms state-of-the-art baselines and significantly improves sentiment recognition accuracy.
    Keywords: multimodal sentiment analysis; text-guided; feature decomposition; contrastive learning; dual cross-modal interaction.
    DOI: 10.1504/IJCSE.2026.10082200
     
  • A new segment routing method with a swap node selection strategy based on deep reinforcement learning for software-defined networks   Order a copy of this article
    by Miao Ye, Jihao Zheng, Qiuxiang Jiang, Yong Wang, Yuan Huang, Ziheng Wang 
    Abstract: The existing segment routing (SR) methods must first determine the routing path and then select swap nodes according to the maximum segment list depth to divide the path into multiple label stacks. This approach lacks flexibility in highly dynamic networks and does not consider the issuance time of the flow table, which affects its routing performance. Therefore, an optimization model for path planning and segmentation strategies that can be simultaneously determined in a software-defined network (SDN) architecture is established herein, and also takes the issuance time of the flow table of the swap nodes corresponding to the segmented paths as one of the optimization objectives. A deep reinforcement learning (DRL) scheme is designed to solve this NP-hard problem. Experiments and their results show that compared with the existing methods, the designed SRP can reduce the time overhead required to complete the SR establishment task while achieving optimal performance metrics.
    Keywords: software-defined network; SDN; segment routing; SR; swap node; deep reinforcement learning; DRL.
    DOI: 10.1504/IJCSE.2026.10082409