In this study, we propose a semi-supervised learning approach for quality-based classification of web services. The increasing volume and diversity of web services necessitate effective methods for automatically categorizing and assessing their quality attributes. Our approach leverages both labeled and unlabeled data, utilizing the labeled data to guide the classification process and the unlabeled data to enhance model generalization. Specifically, we employ a combination of clustering techniques and semi-supervised learning algorithms to assign quality labels to web services based on their functional and non-functional characteristics. Through experiments conducted on a real-world dataset of web services, we demonstrate the effectiveness of our approach in improving classification accuracy and robustness compared to traditional supervised methods. The results indicate that our semi-supervised learning framework not only achieves higher classification performance but also adapts well to varying data distributions and levels of labeled data availability, making it a promising approach for enhancing the automated classification of web services based on quality criteria.

Keywords:

Semi-supervised learning, Web services, Quality-based classification, Unlabeled data, Classification accuracy, Robustness, Data distribution