Using GPT-4 to generate training labels and a domain-adapted BERT encoder, the pipeline classifies review sentiment with 80.8 to 88.6 percent accuracy and no manual annotation.
Examining the relationship between reviews and sales: The role of reviewer identity disclosure in electronic markets,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
support 1representative citing papers
citing papers explorer
-
Lowering the Barrier of Machine Learning: Achieving Zero Manual Labeling in Review Classification Using LLMs
Using GPT-4 to generate training labels and a domain-adapted BERT encoder, the pipeline classifies review sentiment with 80.8 to 88.6 percent accuracy and no manual annotation.