arXiv:2606.25518v1 Announce Type: new Abstract: When people share experiences online, they often express thoughts in two ways: a star rating and a written review. In sentiment analysis, ratings are widely used as convenient weak labels for textual sentiment, yet whether the two actually agree is rarely questioned. This study investigates sentiment-rating incongruence, where the sentiment expressed in review text differs from the sentiment implied by the assigned star rating, in Sri Lankan tourism attraction reviews. A dataset of 16,156 reviews from 2010 to 2023 is analyzed using a transformer-
The proliferation of online reviews and advanced AI sentiment analysis models highlights the discrepancy between explicit ratings and nuanced textual sentiment, making this a timely area of research.
This study is crucial for improving the accuracy of sentiment analysis in AI, which is widely applied in market research, customer service, and social media monitoring, by addressing a fundamental data quality issue.
Understanding behavioral drivers of rating-sentiment incongruence allows for more robust data preprocessing and model training, leading to more reliable AI interpretations of public opinion.
- · AI/ML researchers
- · Customer analytics platforms
- · Businesses relying on user reviews
- · Sentiment analysis software vendors
- · Naive sentiment analysis models
- · Platforms relying solely on star ratings
Improved sentiment analysis models lead to more accurate insights from user-generated content.
Better understanding of customer attitudes and preferences can inform product development and marketing strategies more effectively.
Enhanced AI interpretation of nuanced public sentiment could reduce misinformed decisions by businesses and potentially governments relying on such data.
This signal links to a primary source. Continuum Brief monitors and indexes it as part of the live intelligence stream — we do not republish source content.
Read at arXiv cs.CL