Software is playing the most important role in recent vehicle innovations, and consequently the amount of software has rapidly grown in recent decades. The safety-critical nature of ships, one sort of vehicle, makes software quality assurance (SQA) a fundamental prerequisite. Just-in-time software defect prediction (JIT-SDP) aims to conduct software defect prediction (SDP) on commit-level code changes to achieve effective SQA resource allocation. The first case study of SDP in the maritime domain reported feasible prediction performance. However, we still consider that the prediction model has room for improvement since the parameters of the model are not optimized yet. Harmony search (HS) is a widely used music-inspired meta-heuristic optimization algorithm. In this article, we demonstrated that JIT-SDP can produce better performance of prediction by applying HS-based parameter optimization with balanced fitness value. Using two real-world datasets from the maritime software project, we obtained an optimized model that meets the performance criterion beyond the baseline of a previous case study throughout various defect to non-defect class imbalance ratio of datasets. Experiments with open source software also showed better recall for all datasets despite the fact that we considered balance as a performance index. HS-based parameter optimized JIT-SDP can be applied to the maritime domain software with a high class imbalance ratio. Finally, we expect that our research can be extended to improve the performance of JIT-SDP not only in maritime domain software but also in open source software.

Recently, deep learning frameworks have gained increasing attentions from electroencephalogram (EEG)-based driver’s fatigue estimation, thanks to their unprecedented feature extraction calibre. However, it is still challenging to develop session- and/or subject-independent system, because of the complex structural characteristics of EEG signals. In this regard, this work proposes a novel deep convolutional neural network architecture that can learn spectro-spatio-temporal representation of the vigilance EEG signals, thereby achieving a powerful mental status recognition ability. Specifically, the proposed network pretrained via two novel self-supervision pretext tasks. Further, both differential entropy and EEG signal itself are exploited to acquire rich features. To demonstrate the validity of the proposed methods, this work conduct intra- and inter-subject classification experiments using a publicly available dataset. In the exhaustive experiments, we observed that our proposed framework has practical availability. Specifically, our proposed multiple path structure improved the model’s performance by 3 ~ 7 percentage points (%p) in the session-independent setting and by 6 ~ 9 %p in the subject-independent setting. Besides, the novel self-supervised learning strategy enhanced the performance by 10 ~ 17 and 12 ~ 16 %p in the session-and subject-independent case, respectively. Furthermore, this work also investigate strengths and society-friendliness of our proposed framework.

The increasing adoption of IoT devices on ships and the expansion of platform services present a critical challenge in integrating heterogeneous ship data models into a unified platform data model. The task involves mapping ship Domain-Specific Language (DSL) descriptions to platform indices, complicated by variability and class imbalance. To address these challenges, this paper proposes a framework that combines a sequence-to-sequence model with statistical vectorization techniques. The model generates structured mapping classes, offering flexibility to accommodate diverse equipment and attributes, while training exclusively on connected data mitigates class imbalance. Subsequently, statistical vectorization techniques are applied to identify the correct match among the classified candidates, while ensuring that unconnected data is excluded. This two-step approach enhances recall and guarantees accurate relationships between ship DSLs and platform data indices. The proposed framework is validated using real-world data from 52 ships. Experimental results demonstrate that the sequence-to-sequence model with statistical refinement outperformed single-step and discriminative methods in handling class imbalance and variations when mapping ship DSLs to a unified platform data model. Our method achieved a recall of 89.14 and an F_β-Score of 87.12, which are 4.15 and 1.91 points higher, respectively, than the reference classification method.

Large Language Models (LLMs) demonstrate potential in code generation capabilities, yet their applicability in autonomous vehicle control has not been sufficiently explored. This study verifies whether LLMs can generate executable MATLAB code for software-defined vehicle scenarios, comparing five models: GPT-4, Gemini 2.5 Pro, Claude Sonnet 4.0, CodeLlama-13B-Instruct, and StarCoder2. Thirteen standardised prompts were applied across three types of scenarios: programming-based driving scenarios, inertial sensor-based simulations, and vehicle parking scenarios. Multiple automated evaluation metrics—BLEU, ROUGE-L, ChrF, Spec-Compliance, and Runtime-Sanity—were used to assess code executability, accuracy, and completeness. The results showed GPT-4 achieved the highest score 0.54 in the parking scenario with an overall average score of 0.27, followed by Gemini 2.5 Pro as 0.26. Commercial models demonstrated over 60% execution success rates across all scenarios, whereas open-source models like CodeLlama and StarCoder2 were limited to under 20%. Furthermore, the parking scenario yielded the lowest average score of 0.19, confirming that complex tasks involving sensor synchronisation and trajectory control represent a common limitation across all models. This study presents a new benchmark for quantitatively evaluating the quality of SDV control code generated by LLMs, empirically demonstrating that prompt design and task complexity critically influence model reliability and real-world applicability.