Pilot studies: Using machine learning to analyse radio content in Uganda (2017)
PARTNERS: UN agencies
PROGRAMME OF WORK: The Sustainable Development Goals (SDGs)
PROJECT MICROSITE: http://radio.unglobalpulse.net/uganda/case-studies/
To assess the validity and value of the radio content analysis tool, a prototype developed by Pulse Lab Kampala to analyze public radio content in Uganda, the Lab worked with the government and UN partners on a number of case studies.
The aim of the studies was to understand what type of information can be obtained from radio talk in Uganda and how it might be useful to advance and inform sustainable development and humanitarian action.
By sampling different indigenous languages, types of broadcasters, and locations within Uganda, the pilot studies assessed the potential uses of radio talk across five topics: perceptions towards refugees in Uganda, the impact of small-scale disasters on livelihoods, perceptions on the delivery of healthcare services, understanding the spread of infectious diseases, and monitoring the effectiveness of awareness raising radio campaigns.
Results yielded that relevant information can be extracted from opinions expressed over radio regarding for example the state of healthcare service delivery or the levels of acceptance of host communities towards refugees. Additionally, data analysis of radio content showed it is a valuable source for evaluating the effectiveness of behavioral change campaigns broadcasted on radio.
First hand testimonials on how small-scale natural disasters are affecting the livelihoods of people and communities can also be gauged from public radio discussions and help inform disaster management efforts.
Some instances of public radio data analysis revealed radio may not be the most effective data source to inform on issues relevant to the SDGs, like in the case of informing on the spread of Malaria in Uganda.
The methodology and processes used to develop the radio project as well as the results of the case studies were compiled in a report (to be published). The hope is that these examples will provide inspiration for other innovation projects looking to use radio content and data analytics to inform development and humanitarian efforts.