UN Global Pulse is working with partners to explore how data from social media and radio shows can inform peace and security efforts in Africa. The methodology, case studies, and tools developed as part of these efforts are detailed in a new report entitled: Experimenting with Big Data and Artificial Intelligence to Support Peace and Security.
UN Secretary-General António Guterres underscored in his address to the UN Security Council in December that more countries are experiencing violent conflicts now than at any other time in the past 30 years. Escalating conflicts are destroying progress in education, health, infrastructure, while gains in social cohesion, human rights, and stronger governance are being set back. At the same time, new technological advances, are allowing radical groups to reach an ever-larger audience and exert influence, especially over young people.
The 2030 Agenda for Sustainable Development re-iterates that there can be no sustainable development without peace and no peace without sustainable development. However, new types of war, conflict and violence challenge traditional ways of doing analysis to support peace and security efforts. As highlighted at the Istanbul Innovation Days 2018 (#NextGenGov), emerging global trends are impacting governance mechanisms and collaborative experimentation is needed to address them.
Effective conflict mitigation relies on timely information to identify trends as they emerge and monitor contexts as they evolve. Small data, the type of information obtained from traditional means of analysis, like surveys or focus group discussions, can no longer respond to these emerging, fast-developing challenges. In addition, security constraints in conflict or post conflict areas limit the movement of analysts who have to remain confined in secured locations with reduced capacity to interact with the local population. Analysts also frequently encounter language barriers in these contexts and have limited ability to correct the biases introduced by interpreters and translators.
Using big data and analytics to fill information gaps
Big data refers not only to the large volumes of data now available, but also to the accompanying processes and technologies for collecting, storing, and analyzing it. In recent years, technology has allowed for the expansion of crowdsourcing for social activism and citizen journalism. For example, citizen-reporting projects use tools like Ushahidi (which translates to testimony) or OpenStreetMap to provide real-time reporting and mapping.
At the same time, businesses increasingly are mining the digital trails we leave behind to predict consumer behaviour, track emerging trends in the market, and monitor operations in real time to improve sales and profit margins.
For sustainable development and humanitarian practitioners, big data and new technologies hold great potential to help measure the effectiveness of projects and programmes, and proactively adjust their implementation based on the realities on the ground.
Using techniques similar to those developed by private sector, UN Global Pulse has been working – through its lab in Kampala – on using big data and AI to support achievement of the SDGs in Africa. Over the last three years, the Lab has been working with partners to explore uses of data from social media feeds and radio shows, among others, to understand what people think, what their needs are, how policies are being implemented, or the effects of humanitarian crises on populations and early warning response.
More recently, the team has been experimenting with these two data streams – mining social media in Somalia and analysing people’s voices from radio shows in Uganda – to understand how they can inform SDG16: Promote Just, Peaceful, and Inclusive Societies.
The first test case used data extracted from social media, namely posts from public Facebook pages and groups, to analyse how influencers and fake news might be shaping discussions among online users in Somalia and to identify trending topics relevant to SDG16.
The second test case analysed public discussions on radio to detect instances of rumours and misconceptions, and of social tensions as reported by listeners in Uganda. The analysis of people’s voices was done using an Automated Speech Recognition (ASR) toolkit that Pulse Lab Kampala developed and which uses convolutional neural networking technology to convert speech to text.
The methodology and findings of the experimentation process, together with the software developed as part of the two test cases, are outlined in the new report: Experimenting with Big Data and Artificial Intelligence to Support Peace and Security.
The report also details the functionalities of QataLog, a new tool that UN Global Pulse is developing to help extract, analyse and visualize data. It currently allows users to extract useful information from social media and radio and to analyse it for topics of interest, such as humanitarian action or peace and security, using a combination of manual annotations and automatic processes that include translation, geo-location and text classification. Users can visualize the insights over time and territories and can download the raw data for further analysis. The tool is being tested with several UN teams and will be continuously refined.