External Events

Balance: Facebook Project Journalism Hackathon

Share this post:

Learn how two emerging technology specialists created ‘Balance’ – a news service which provides a cross publisher view on similar and opposing articles.

Anna Thomas and Dan Cunnington from Emerging Technology attended the event in London and the day kicked off with a panel discussion from various companies within the media industry. The panel included Jon Heaton – Partner Engineer at Facebook, Yusuf Omar – Senior Social Reporter at CNN, David Milne – Head of Digital Publishing at STV and Andrew Mullaney – CTO of Newswhip. The panel discussed upcoming trends within the media industry, such as emerging storytelling formats and live streaming. It was really interesting to hear thoughts around the use of Virtual and Augmented Reality and how that will affect publishers in terms of the ability to analyse, process and display content to users. This highlighted the need for new ways of exploring news content, moving away from the traditional article based format. Fake news was a hot topic, with all members of the panel explaining the scale of the problem and also the technical difficulty.

Teams then began to brainstorm around the challenges discussed and it was here that ‘Balance’ was born. Fake news is a big problem and users are losing the ability to trust pieces of content they read online or in the media. Users can often be influenced by articles that play on emotional triggers or portray certain entities/characters in a certain way. For an end user, It can be difficult to see a balanced view of the news surrounding a particular topic and news platforms such as Facebook wish to remain impartial on the content they display. ‘Balance’ aims to provide a cross publisher, cross platform view of news content related to a specific topic. As the hackathon coincided with the day Theresa May officially triggered Article 50, we focused our prototype on the topic of ‘Brexit’ in order to collect recent content. ‘Balance’ captures social data from CrowdTangle (http://www.crowdtangle.com/) and articles from the IBM Watson Discovery News Service (https://www.ibm.com/watson/developercloud/discovery-news.html ). Using the traditional article based format alongside emerging social formats allows a broad view to be generated of news content surrounding a particular topic.



With news data flowing from two services, we then analysed the entities and emotional sentiment of the articles using the IBM Watson Alchemy Language (https://www.ibm.com/watson/developercloud/alchemy-language.html ) service, taking inspiration from Zoe Lanham and Dominik Ullmann with their Trump vs Clinton comparison at the ETS December hackday. This allowed us to identify certain entities or topics and have a sentiment score associated with that topic for ‘Anger’, ‘Fear’, ‘Disgust’, ‘Joy’ and ‘Sadness’ emotions. If an article is in favour of Brexit, we’d expect to see a negative score for ‘Disgust’ and ‘Sadness’ as well as a positive score for ‘Joy’. With this data, we can then recommend similar and also opposing articles to give the user a balanced view of the news.



This helps to reduce the impact of fake or manipulative news, as the user can see the full picture on both sides of an argument and it also helps to discover new content from sources they would not normally follow.

To build the app we used Node-RED (a visual programming tool to quickly wire together different API’s http://nodered.org/ ) and also Angular.js to display the data in a visual and meaningful manner.

To check out the demo, please visit the following URL: http://balance.eu-gb.mybluemix.net/



More External Events stories

Finding Hidden Cell Phones in Prisons

Myself and some colleagues from IBM’s Emerging Technology attempted a two-day project just before Christmas this year, because we thought we might have the tech to tackle a big problem: cell phones in prisons, we know prisoners have them, but they are rarely ever tracked down. This is something the UK government is currently planning to spend […]

Continue reading

Building the IBM Emotive Droid

Animating emotion. This is a project to show of the use of affective computing (Emotional AI) in the Watson suite, which is IBM’s collection of cloud-based AI API’s. We hooked this into The Waston SDK for Unity, which allowed us to use this as a 3D Environment. The goal here was simple: to create an […]

Continue reading

Writing a Chess Game with IBM Watson, and Unity

With the release of the Watson Unity SDK in 2018, myself and Amara Graham (Keller) set out to build a chess game that could be completely voice controlled. In order to tackle this, we had three major tasks ahead of us, each of which had a clear solution. First of all, we would need our application to […]

Continue reading