BERT-Based Deceptive Review Detection in Social Media: Introducing DeceptiveBERT
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
In recent years, the Internet has facilitated the emergence of social media platforms as significant channels for individuals to express their thoughts and engage in instantaneous interactions. However, the reliance on online reviews has also given rise to deceptive practices, where anonymous spammers generate fake reviews to manipulate the perception of a product. Ensuring the integrity of the online review system requires identifying and mitigating fake reviews. While existing machine learning (ML)- and neural network (NN)-based sentiment analysis methods can detect deceptive reviews, they often suffer from long training times, high computational resource requirements, and memory constraints. This study aims to overcome these limitations by introducing a transformer-based “deceptive bidirectional encoder representations from transformers (DeceptiveBERT) model.” This model utilizes contextual representations to enhance the precision of deceptive review identification. Transfer learning is employed to leverage knowledge from a pre-existing BERT base-uncased word embedding model, enabling efficient feature extraction. The proposed model incorporates a combination of classification layers to categorize reviews into two distinct categories: deceptive and truthful. Additionally, the study addresses the challenge of imbalanced datasets by utilizing three separate datasets and implementing appropriate methodologies for dataset curation. The effectiveness of the DeceptiveBERT model was evaluated through experimentation. The results demonstrate its efficacy, with the model achieving accuracy rates of 75%, 84.79%, and 81.08% on the Ott, YelpNYC, and YelpZip datasets, respectively.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it