Submisison Type: Talk + Q&A (30 mins)

Understanding public sentiment is critical for brands, government agencies, and organizations seeking to make informed decisions. This process, known as social listening, was traditionally labor-intensive, involving manual annotation and review of thousands of online opinions. Conventional sentiment analysis tools, such as VADER, often struggled with nuanced language, sarcasm, and context, leading to unreliable insights and a continued dependence on human involvement.

The advent of large language models (LLMs), such as Google’s Gemini, is transforming this space by automating thematic and sentiment analysis with remarkable accuracy. Thanks to their large context window and advanced natural language understanding, LLMs can decode nuanced public opinions, detect sarcasm, and extract actionable insights at scale. This automation reduces human involvement, accelerates results, cuts costs, and democratizes access to powerful sentiment analysis tools for individuals and brands alike.

In this session, we’ll explore how LLMs are reshaping public perception mining and examine real-world use cases demonstrating their impact on social listening, from crisis management to brand monitoring and public policy evaluation.

Key Takeaways:

- How LLMs improve sentiment analysis by detecting nuance, sarcasm, and context.

- The role of LLMs in automating public perception mining at scale.

- Practical strategies for using LLMs in social listening and sentiment tracking