This repository contains the dataset and code for our paper.
All data is available at all_data.
Source label 0 represents real hotel reviews and label 1 represents fake/ LLM-generated hotel reviews.
Topic Modeling features can be accessed interactively in topic_analysis
All generation code is available at LLM_generation.
XLM-Roberta, Random Forest and Naive Bayes models, together with interpretable features are available at Deception Detection Models.