Text Classification Using Machine Learning Methods, Applications, and Future Directions in Secure Cloud-Native Healthcare Analytics

Abstract

Text classification using machine learning has emerged as a critical component in healthcare analytics, enabling the automated interpretation and categorization of large volumes of unstructured clinical and administrative text. This paper presents a comprehensive overview of machine learning–based text classification methods, their applications, and future research directions within the context of secure cloud-native healthcare analytics. We examine traditional approaches such as Naïve Bayes, Support Vector Machines, and decision trees, alongside advanced deep learning models including convolutional and recurrent neural networks, as well as transformer-based architectures like BERT. The study highlights how cloud-native, API-enabled architectures enhance scalability, interoperability, and real-time processing of healthcare data while addressing security and privacy requirements. Key healthcare applications such as clinical document classification, medical coding, sentiment analysis of patient feedback, and disease surveillance are discussed. Furthermore, the paper analyzes challenges related to data privacy, interpretability, model bias, and computational efficiency in cloud environments. Finally, future directions are outlined, including secure federated learning, explainable AI, and resource-efficient models, which are expected to play a pivotal role in advancing trustworthy and scalable healthcare text analytics.