A review of explainable semi-supervised methods in multivariate time series analysis

Recent advances in information and communication technologies have led to the widespread use of smart meters, wearable sensors, and other devices in healthcare, industry, and transportation that These systems generate large volumes of sequential data, yet fully annotating them is both expensive and often impractical. As a result, only a small portion is usually labeled, which limits the use of supervised learning and make unsupervised methods less effective as they ignore the valuable information in those labels. This has led to growing interest in semi-supervised methods, which can learn from both labeled and unlabeled data. However, many such methods rely on complex black-box models, making their decision processes opaque-particularly problematic in high-risk domains like medicine and finance. Explainable AI (XAI) has therefore become essential for building trust and ensuring accountability in these settings. This review surveys recent advances in explainable semi-supervised methods for multivariate time series analysis. We introduce a taxonomy based on how explainability is integrated, categorizing methods as white-box (transparent), intermediate, and post-hoc (opaque). Within each category, we further classify them by their most distinctive characteristics: model class for white-box, interpretability integration strategy for intermediate, and explanation technique for post-hoc. We also discuss common practices in handling partial supervision and model interpretability, and highlight key challenges in sequential data analysis-such as the choice of performance metrics and explanation techniques. Overall, explainable semi-supervised methods represent a promising direction for future research, offering tangible benefits across a wide range of real-world time series problems, particularly in high-stakes settings with limited labeled data.