Needs Companion: A Novel Approach to Continuous User Needs Sensing Using Virtual Agents and Large Language Models

Area of Science:

• Human-Computer Interaction
• Artificial Intelligence
• Service Science

Background:

• Services are crucial for daily life, necessitating personalized approaches.
• Current methods often infer needs from behavior logs, lacking direct detection.
• Creating a human-centered society requires understanding individual service requirements.

Purpose of the Study:

• Introduce Needs Companion for automatic individual service need detection.
• Establish a foundation for accurate needs sensing.
• Provide interpretable data for personalized service development.

Main Methods:

• Developed a needs data model based on the 6W1H framework.
• Employed virtual agents for needs elicitation.
• Utilized large language models (LLMs) for automated needs analysis and extraction.

Main Results:

• The Needs Companion system demonstrated accurate and rapid detection of individual service needs.
• Experimental results validate the system's effectiveness in identifying user requirements.
• Generated interpretable data crucial for tailoring services.

Conclusions:

• Needs Companion advances automated needs detection for personalized services.
• The research contributes to machine learning, human-centered design, and requirements engineering.
• This work paves the way for more responsive and user-centric service systems.