Research on Text Mining focuses on developing advanced methods to extract, analyze, and interpret information from large volumes of textual data across diverse domains. Recent studies highlight frameworks for automated generation of research workflows using Positive-Unlabeled learning with SciBERT, Flan-T5, and ChatGPT to structure and visualize workflow stages. Applications in environmental research reveal AI-driven trend analysis and thematic exploration, while healthcare-focused text mining improves diagnostic accuracy and supports personalized medicine. Additionally, advancements in feature selection and extraction techniques enhance the efficiency and effectiveness of text mining processes. Overall, these innovations integrate machine learning and AI tools to enable more accurate, scalable, and insightful analysis of textual data.