Using Machine Learning to Accelerate Sustainable Development Solutions in Uganda
September 14, 2017
A year and a half after it was prototyped, the radio content analysis tool developed by Pulse Lab Kampala and partners has become fully operational. The findings and lessons learned during the process were compiled in a report entitled: “Using Machine Learning to Analyse Radio Content in Uganda - Opportunities for Sustainable Development and Humanitarian Action.”
The recent Artificial Intelligence (AI) for Good Global Summit has brought together partners to define a roadmap for governments, industry, academia, media, and civil society to develop AI in a safe, responsible and ethical manner benefiting all segments of society. At the summit, the radio content analysis tool was showcased as one of the applications of AI currently in use at the UN.
The tool was designed to leverage public radio content as a source of information to inform on issues relevant to sustainable development. The most complex part in the development of the prototype is capturing the transcription of spoken words into written text. This technology, called speech recognition, is used in applications ranging from simple voice dialing (e.g. "Call home") to fully automatic speech-to-text processing where every word is being converted into text (e.g. dictation to a document or email).
The world’s largest IT companies, including Apple, Google, Microsoft and IBM, invest significant resources in speech recognition for their products. There are also companies that specialise in speech recognition as Nuance Communications (Apple’s supplier) or HTK. This type of companies offer automatic speech-to-text dictation in about 50 languages, but languages and dialects from the African continent are not available among them.
The radio content analysis tool was developed as part of a project conducted by Pulse Lab Kampala in collaboration with the Stellenbosch University in South Africa. The tool works by converting public discussions that take place on radio in various African languages into text. Once converted, the text can be searched for topics of interest. The tool is now fully functional in the Northern and Central regions of Uganda and available for three languages: Luganda, Acholi and English (as spoken in the country).
The report outlines the methodology and processes of the radio content analysis tool, distills the technology behind its creation and presents the lessons learned along the way. It also details the results of several pilot studies that were conducted together with partners from the Government, UN agencies and academia to understand the validity and value of unfiltered public radio discussions for development.
The hope is that the processes and lessons detailed in the report can serve as examples and inspiration for using radio talk and data analytics to inform decision-making processes in development and humanitarian scenarios, in contexts where other sources of data may be missing or insufficient.
Using Machine Learning to Analyse Radio Content in Uganda from Global Pulse
Uganda’s population is the youngest in the world, with 77% of its population being under 30 years of age. The country is now gaining international recognition for the development of Artificial Intelligence products by its youth.Listen to insights from the young Ugandans working at Pulse Lab Kampala on the development of the radio content analysis tool.
Cross-posted from the United Nations Global Pulse Blog.
The stories behind the numbers in Kivu
June 10, 2016
Results, results, results. The age old monitoring and evaluation question: how do you [actually] draw a connection between transformational changes in the lives of people and the development projects that aim to help them?
The hard part is that the traditional monitoring approach does not focus on measuring outcome indicators, a weakness corrected by a new monitoring method: SenseMaker Narrative Capture. This initiative focuses on transformational changes, and uses qualitative and quantitative methods and collects narratives shared by the beneficiary populations.
As head of the Monitoring and Evaluation unit in the UNDP Democratic Republic of Congo country office, I led the implementation of this new monitoring and evaluation approach in South Kivu. Overall, the project was designed to to support the stabilization of the South Kivu region, which has been part of a conflict since 1994 among several actors looking to expand their territories in the Great Lakes Region.
Overall we believe that strengthening community management of conflict resolution and social infrastructure will help reduce potential sources of tension, which will help displaced and refugee populations return and reintegration process.
Monitoring change with a participatory approach
Generally, we were interested to learn about the changes in the life of communities involved in this joint programme developed by UNDP, UNICEF and FAO and particularly, we wanted to capture people’s experiences and feelings around the Kivu conflict, peace-keeping efforts surrounding the conflict, and the reintegration experiences of displaced individuals.
For this purpose, we approached different organizations and community leaders involved in the peace process following the conflict in the region. Our idea was to seek for their support designing monitoring tools and instruments we were planning to use and, because they took part in this first phase of the process, the tools obtained added value to the project. This participatory approach ensured that the content of the tools and questionnaires was well aligned with the reality in the field. This reality check empowered us to move to the most challenging part of the process, the data collection.
Capturing the stories behind the data
During the the data collection process, more than a thousand community members shared with us their story about the conflict, the stabilization and the peace process.
On this process of capturing the stories, what mostly amazed us, beyond their content, was the storytellers’ feedback:
“By sharing this story I realize how was my life before, during and after the conflict, I realize how bad a conflict can be, why it is important to live as a community, to bring our children up with a new mindset. I realize how the different actors: the local authority, the church, the national army, the self-defense groups were interacting to either maintain crisis situation or to improve the situation of the communities”.
Some of the participants also shared their positive feedback on the way the data collection was done:
“The way you designed the questionnaire without asking me to share my opinion but to tell my story was fantastic. I used to give my opinion for surveys conducted by other organizations but I was never able to look back on the conflict and all the horror, the death, the tears, the food insecurity that we had to face everyday.”
Through this methodology, we realized that assessing the situation helps the storytellers focus not only on their opinion but also on their past experience. That is why we believe that Sense@Maker is an interesting and relevant addition to the M&E exercise as it is a realistic tool based on the commitment and strong participation from the beneficiaries and we plan to use it to influence future programme design and implementation.
Among the findings, one pointed out that education is a top concern for the communities. According to the results, communities find education a key component to promote skills, knowledge and new employment opportunities. So we are currently studying how education can be used to achieve a deeper impact in shaping attitudes towards conflict resolution and expanding access to social services. We will keep you in the loop!