Data-Driven Persona Generation via Structured Analysis and LLM Prompt Injection: Comparing Analytical Method Selection and Combination Strategies

Abstract

Consumer personas are essential for user understanding and marketing strategy, yet manual construction remains costly and difficult to scale. We propose a data-driven persona generation framework that systematically compares three interpretable text mining methods-SNA, LDA, and K-Means-across eleven experimental configurations and injects the resulting structured representations into LLM prompts. Experiments on two consumer review datasets (VOC: n=965; VAD: n=2,370) with N =20 repeated trials sh

Categories

Framework mappings

Suggested from the entry's categories.

@misc{kim2026datadriven,
  title = {{Data-Driven Persona Generation via Structured Analysis and LLM Prompt Injection: Comparing Analytical Method Selection and Combination Strategies}},
  author = {Yujin kim and Jaekwang Kim},
  year = {2026},
  doi = {10.2139/ssrn.7281659},
  url = {https://doi.org/10.2139/ssrn.7281659}
}