Enhancing products performance evaluation through hybrid DistilRoBERTa and BiGRU models
Shoukat Ullah, Aurangzeb Khan, Aman Ullah, Muhammad Nawaz Khan, Rehan Tariq Chohan, Farhan Aadil, Khair Ullah Khan · 2026
Read Enhancing products performance evaluation through hybrid DistilRoBERTa and BiGRU models by Shoukat Ullah, Aurangzeb Khan, Aman Ullah, Muhammad Nawaz Khan, Rehan Tariq Chohan, Farhan Aadil, Khair Ullah Khan (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/pmc-13210267.txt.
Description
Star ratings hide how hard a product is to live with, and a 2026 study headed by Shoukat Ullah builds a model that reads reviews for complexity instead. The research article pairs a transformer language model with a recurrent network to judge Amazon reviews along two appraisal dimensions: user-friendliness and judged effectiveness.
Human annotators first labeled reviews by perceived complexity, and an explanation method cross-checked those labels for consistency. The hybrid model then learned both context and word order, reaching higher cross-validated accuracy than the forest, transformer-only, and large language model baselines cited. Built-in explanations show which words drive each judgment, making the scores auditable for shoppers and sellers.
Catalog record
| CREATOR | Shoukat Ullah, Aurangzeb Khan, Aman Ullah, Muhammad Nawaz Khan, Rehan Tariq Chohan, Farhan Aadil, Khair Ullah Khan |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 65 KB |
| VERSION | 1.0 |
| LICENSE | CC BY 4.0 |
| RIGHTS | CC BY |
| SOURCE | Europe PMC |