Can using machine learning and natural language processing help more citizens get their views heard on digital democracy platforms?
The experiment will test whether machine learning and natural language processing (NLP) can effectively cluster similar proposals and recommendations from ‘like minded citizens’ on the Consul digital democracy platform. This should make it easier for citizens on the platform with shared interests to find each other and work together. The experiment will generate insights on whether this increases the number of citizen proposals that receive sufficient support from other citizens to trigger further action by local governments.
NLP allows computers to understand, interpret and extract key information from human language.
Around the world digital democracy platforms are engaging new groups of people and empowering citizens to contribute to policy making. As trust in traditional democratic institutions declines, deliberative platforms offer a way to build new relationships and trust between citizens and policy-makers. At the moment, however, only small numbers of proposals from people on the Consul platform receive the minimum number of support votes that are required for further action from the local government. This is due to the large volume of content on the platform, with ideas often spread over dozens of different proposals from citizens. This experiment will test how machine learning and NLP can overcome the current information overload and help people use digital democracy platforms to greater effect.
The experiment will help us understand how to better connect like-minded citizens and similar ideas to enhance collective action. As well as being relevant for all types of digital democracy initiatives, it will have broader relevance for social movements, labour organising, and initiatives that aim to crowdsource innovation.