Statistical Foundations for a Google Play User-Review Sentiment Index: Signal Fusion, Shrinkage, Distributional Validation, and Dynamic Smoothing
Researchers define a Google Play review sentiment index using signal fusion, shrinkage, and Kalman smoothing.
The paper defines a Google Play review sentiment index that treats star ratings and text-sentiment scores as noisy measures of latent valence and fuses them with covariance-aware inverse-variance weighting. Aggregates use bounded helpfulness and recency weights, then shrink toward a population mean by estimated precision rather than a review-count cutoff. Rating histograms serve as a distributional check across API sort orders, and a local-level Kalman filter produces a denoised trend. Proofs cover BLUE estimation, Gaussian shrinkage, empirical-process results, the delta method, and exact Gaussian Kalman filtering.
- Inverse-variance weighting fuses star ratings with text-sentiment scores.
- Review estimates shrink toward a mean using estimated precision.
- Histograms diagnose samples from different Play API sort orders.
- A Kalman filter supplies the denoised temporal sentiment trend.
- Vendors
- Products
- Google Play
From the original article
1 min read at arxiv.orgWe develop a statistically explicit sentiment index for Google Play user reviews and establish the mathematical results supporting its construction. Normalized star ratings and text-sentiment scores are treated as noisy measures of latent review valence and fused by covariance-aware inverse-variance weighting.
By Marco Mandap · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.