Utilizing Large Language Models to Synthesize Product Desirability Datasets

TL;DR

This paper demonstrates how large language models like GPT-4o-mini can efficiently generate synthetic product reviews that maintain high sentiment accuracy and diversity, aiding in scalable product desirability testing.

Contribution

The study introduces three novel methods for synthesizing product reviews using LLMs, highlighting their effectiveness and cost-efficiency for creating valuable testing datasets.

Findings

High sentiment alignment with Pearson correlations 0.93-0.97

Supply-Word method achieves highest diversity and coverage

Synthetic data provides scalable, cost-effective testing solutions

Abstract

This research explores the application of large language models (LLMs) to generate synthetic datasets for Product Desirability Toolkit (PDT) testing, a key component in evaluating user sentiment and product experience. Utilizing gpt-4o-mini, a cost-effective alternative to larger commercial LLMs, three methods, Word+Review, Review+Word, and Supply-Word, were each used to synthesize 1000 product reviews. The generated datasets were assessed for sentiment alignment, textual diversity, and data generation cost. Results demonstrated high sentiment alignment across all methods, with Pearson correlations ranging from 0.93 to 0.97. Supply-Word exhibited the highest diversity and coverage of PDT terms, although with increased generation costs. Despite minor biases toward positive sentiments, in situations with limited test data, LLM-generated synthetic data offers significant advantages, including scalability, cost savings, and flexibility in dataset production.