Institute of Information Science, Academia Sinica

Recent advances in generative AI and large language models (LLMs) have opened up new possibilities for text analysis research. From data collection and linguistic annotation to content analysis, AI is increasingly becoming an important tool for assisting researchers throughout the research process. However, questions remain as to whether AI can genuinely understand linguistic meaning and how reliably it can perform different text analysis tasks. This talk introduces the use of AI for data collection, linguistic annotation, and textual analysis. It also discusses the challenges and methodological considerations that researchers may encounter when incorporating AI into these processes. Drawing on several research cases, including hate speech detection, disaster-related information filtering, suicidal ideation detection, and sponsored-content identification, the talk examines both the capabilities and limitations of generative AI across different text analysis tasks.