Neural Network Models to Study Framing and Echo Chambers in News

An interesting study is to construct a model of the media representations of the world, considering features from social discourse such as crime, race, and so on. In other words, we generate “interpretive frames” that introduce selected media biases and predispose the system to look at the world in a certain way. Therefore, any data that is presented to this system will be viewed through this lens where certain outcomes are anticipated, and the communicative effects will depend on the associated inferences.

We propose a few models and studies:

• Model 1: Generating different summaries of a news story by opinion spectrum based framing
• Model 2: Generating parody news stories based on an image using neural storytelling
• Model 3: Simulating evolving window of acceptability
• Model 4: Simulating the spread of unverified information
• Study: News recommender systems causing filter bubble effect

With the help of these, we may conclusively demonstrate real world phenomena that highlight issues of media bias, framing, echo chambers, and the wisdom of the crowd. Such an analysis may shed some light on the political climate and how crowd consensus comes to be.