# Beyond Stars: Bridging the Gap Between Ratings and Review Sentiment with LLM

arXiv: https://arxiv.org/abs/2509.20953 · HTML: https://mlsanity.com/paper/2509.20953/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 21 references that are in the map.
- First posted on arXiv 25 Sep 2025.

## Builds on (most cited first, 10 of 10)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Retrieval-Augmented Generation for Large Language Models: A Survey](https://mlsanity.com/paper/2312.10997/) (2023)
- [Explainability for Large Language Models: A Survey](https://mlsanity.com/paper/2309.01029/) (2023)
- [Fault in your stars: An Analysis of Android App Reviews](https://mlsanity.com/paper/1708.04968/) (2017)
- [Comparative Sentiment Analysis of App Reviews](https://mlsanity.com/paper/2006.09739/) (2020)
- [Emerging App Issue Identification via Online Joint Sentiment-Topic Tracing](https://mlsanity.com/paper/2008.09976/) (2020)
- [How Effectively Do LLMs Extract Feature-Sentiment Pairs from App Reviews?](https://mlsanity.com/paper/2409.07162/) (2024)
- [A Comparison of Automatic Labelling Approaches for Sentiment Analysis](https://mlsanity.com/paper/2211.02976/) (2022)

## Abstract (arXiv)

We present an advanced approach to mobile app review analysis aimed at addressing limitations inherent in traditional star-rating systems. Star ratings, although intuitive and popular among users, often fail to capture the nuanced feedback present in detailed review texts. Traditional NLP techniques -- such as lexicon-based methods and classical machine learning classifiers -- struggle to interpret contextual nuances, domain-specific terminology, and subtle linguistic features like sarcasm. To overcome these limitations, we propose a modular framework leveraging large language models (LLMs) enhanced by structured prompting techniques. Our method quantifies discrepancies between numerical ratings and textual sentiment, extracts detailed, feature-level insights, and supports interactive exploration of reviews through retrieval-augmented conversational question answering (RAG-QA). Comprehensive experiments conducted on three diverse datasets (AWARE, Google Play, and Spotify) demonstrate that our LLM-driven approach significantly surpasses baseline methods, yielding improved accuracy, robustness, and actionable insights in challenging and context-rich review scenarios.