Cyber hate speech on twitter: Analyzing disruptive events from social media to build a violent communication and hate speech taxonomy
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
The attack against the Charlie Hebdo weekly in Paris, in the year 2015, was a disruptive event that generated an important public reaction in social networks, creating the opportunity to study the phenomenon of violent communication and hate messages on Twitter. In the days after the attack (between January 7 and January 12), a sample of more than 255,000 tweets with the hashtags #CharlieHebdo, #JeSuisCharlie and #StopIslam was collected. An analysis was made using qualitative and quantitative approaches to contrast the level of agreement between the different methods used. In the first place, messages were classified as tweets that contained violent and hate speech or general messages, following the inclusion criteria that based on experience and the scientific literature were defined by the Principal Investigator. Then, three pairs of judges classified the sample using the excluding criteria previously defined, according to which ten types of violent speech communication were identified, which were reduced to five essential categories. After the qualitative analysis, the methods of Data Mining were used with the purpose of extracting systems of rules for the classification of the type of speech, beginning with 18 variables derived from each tweet, including date, favorites or the type of software used for the tweet, among others. The results show that disruptive events are followed by communications that show spatial temporal and textual patterns clearly identifiable; this allows the authors to propose a methodology to classify in a very precise way, those messages that contain hate or violent speech.
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 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 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