Sentiment-based Classification of Radical Text on the Web
Bibliographic record
Abstract
The total number of webpages has grown substantially since the birth of the Internet. So too have the number of webpages dedicated to radical yet subtle content. As these new circumstances have necessitated a guided data collection method, one that can sidestep the laborious manual methods that have been classically utilized, simple keyword analysis has not been sufficient to identify radical sites on Web 1.0 - pro-extremist, anti-extremist, and news sites, for example, may use the same keywords to discuss the same event but have a very different motivation. In an effort to explore this problem, we completed an exercise involving the use of a web-crawler to collect 20,000 webpages from five sentiment-based classes to assess their differences: (1) radical Right sites, (2) radical Islamic sites, (3) anti-extremist sites, (4) news source sites discussing extremism, and (5) sites that did not discuss extremism. Parts-of-Speech (POS) tagging was used to identify 198 of the most frequent keywords within the data, and the sentiment value for each of these keywords was calculated for each webpage using sentiment analysis. With these values, a decision tree was applied to three classification models. Results suggest that radical Islamic text can be classified at a much higher rate of success than radical Right text.
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 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 | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.001 | 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 itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, 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".