Enhancing fake news detection through estimating user tendencies to spread fake news
Bibliographic record
Abstract
The growing influence of social media on how people consume information has reshaped the landscape of public communication. Alongside its benefits, this shift has led to the faster spread of fake news, reducing public trust and influencing people’s perception of events. Gaining insight into how fake news propagates and understanding the roles different users play in its dissemination are essential steps toward effective detection. In this research, we investigate how predicting users’ sharing behaviors can improve fake news detection (FND). We introduce a regression-based approach to estimate a user’s Tendency to Spread Fake News (TSFN) by leveraging linguistic features derived from their online posts. To train and evaluate the model, we present two new datasets, each comprising 5000 users. Subsequently, we employ the trained TSFN estimator models for the detection of fake news, presenting a two-step FND system. In the first step, for a given news item, the system estimates the TSFN scores of its spreaders using the trained estimators. Then, leveraging these scores, the system determines the authenticity of the news item. By further combining news content features, the system achieves improved performance. Experimental results indicate that the proposed framework performs reliably even in the early stages of news dissemination. Moreover, we explore how emotional signals contribute to distinguishing between fake and real news and to identifying fake news spreaders, offering valuable insights into the models’ decisions. • Introduce a regression model to estimate users’ tendency to spread fake news. • Propose a two-step system leveraging user tendencies for fake news detection. • Demonstrate effectiveness in detecting fake news even in early propagation stages. • Integrate news content features to improve detection performance. • Analyze emotional signals’ role in identifying fake news and its spreaders.
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.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.017 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.001 |
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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".