In project we are studying the propagation of unreliable content across Wikipedia projects. Specifically, we identify articles annotated with content reliability issues in a given language, and analyze how this contest project to other projects.
Description
Description
| Status | Subtype | Assigned | Task | ||
|---|---|---|---|---|---|
| Resolved | diego | T288333 Understanding the spread of disinformation on Wikipedia | |||
| Resolved | paramita_das | T305390 Cross-Lingual Article Quality Exploration |
Event Timeline
Comment Actions
Updates
- We are analyzing how the speed of propagation changes when a given article in English is tagged with a content reliability template.
Comment Actions
Updates
- We are applying Agglomerative Clustering to compare different propagation path, considering language, topic and reliability issues.
Comment Actions
Updates
- We have found an interesting behavior for content that has been signaled as unreliable. That content tends to spread faster to other projects after been marked with reliability-related template.
Comment Actions
Updates
- We are studying a new method for sequence prediction, using Temporal LSTM.
- At the end of September we are planning to start writing the paper with our main findings.
Comment Actions
Updates
- Previous studies were based on content that has been tagged as unreliable in English, and spread to other languages. Now, we adding 10 new languages. This will help us to differentiate different sources of unreliable content.
Comment Actions
Updates
- We are exploring new features for predicting content propagation, such as countries mentioned in items, and some geographical characteristics.
Comment Actions
Updates
- We are summarizing the results. In the following weeks we should have a report on this.
Comment Actions
Updates
- We discovered a relationship between topics and propagation probability.
- We have improved the propagation prediction algorithm.
Comment Actions
Updates
- We are testing a new DL model to predict content propagation, using content reliability as one of the features.
Comment Actions
Updates
- We are shaping the paper and checking which new experiments would be required.
Comment Actions
Updates
- I'm currently studying the propagation of "climate change" related items (thanks Isaac for the dataset)
Comment Actions
Updates
- I'm preparing a short presentation with the main findings of this project. I'm planning to present these results during the next disinformation working group meeting.
Comment Actions
Updates
- We are exploring cross-lingual article quality. Details on this task: T305390
Comment Actions
@diego can this task be resolved? If not, please share an update about the possible next steps and if you don't intend to prioritize it in the coming 6 months move it to Research-Freezer .