After more than a year of setting up and structuring, the Grain network has been officially launched. The launch took place on Monday 20 November 2023, during a webinar organised on the Zoom platform. The event was attended by all the members of the consortium responsible for running the network (IPAR, CSEA and Sunbird AI). Also taking part in the meeting were the beneficiaries of the call for proposals on the theme of "Improving inclusion and gender equality through artificial intelligence in sub-Saharan Africa as well as experts in gender equality and the AI ecosystem in Africa, notably the AI4D network, to explore the genesis of the network and reflect on avenues for sustainable collaboration.
The meeting opened with a presentation of the GRAIN project, including its research objectives and prospects for sustainability. The member institutions of the GRAIN project consortium were also introduced, followed by a presentation of the GRAIN network. AI4Dthe initiator of the project.
In addition to the official launch of the network, the aim of the meeting was to create a platform for exchange between stakeholders to promote a better understanding of gender equality in AI. Introducing GRAIN, Dr Laure Tall of the IPAR called on members to "make the network sustainable and sufficient to harness knowledge to promote learning and innovation".
With the aim of broadening its social base, in October 2022 the GRAIN network launched a call for proposals inviting AI players to submit projects addressing the use of AI to tackle issues related to gender inclusion and equity in Sub-Saharan Africa. The launch of the network was also an opportunity for the nine beneficiaries from African universities and research institutions to present their projects. In the various speeches, the team leaders highlighted the many challenges facing Africa in terms of gender balance and AI to propose research topics.
The barriers are numerous and include insufficient gender-disaggregated data, limited skills and low participation in the AI economy, a dearth of quantitative and qualitative research, barriers to gender digital equality, the inclusion of women, limited access for women in technology, language issues, the influence of social structures on women's trajectories in innovation and STEM fields, etc.
Dr Olayinka Jelili Yusuf of the University of KWASU based in Kwara State, Nigeria, team leader of the "AI for women in agriculture" project e - AI4WIA " said in his presentation that their project responds to the challenges of climate change, which is leading to a decline and instability in agricultural production results, particularly for women farmers. He also pointed out the socio-cultural limitations in terms of resource ownership and the lack of predictive weather information needed to make informed decisions about the optimum cropping calendar. The study then seeks to propose an AI-based solution to the threat - the double whammy of worsening climate change and dependent rain-fed agriculture.
Another study entitled "AI for Women in Aquaculture" led by Mutiat Mohammed from Summit University, Offa based in Kwara State, Nigeria, targets women in aquaculture with limited access to training, tools and markets. The study highlights the lack of a platform for coordinating efforts and disseminating information, and the timid adoption of artificial intelligence and Industry 4.0 technology, among other things. In this way, research will adopt artificial intelligence techniques in aquaculture and ensure gender equity, diversity and inclusion, while minimising unintentional bias and providing many opportunities for women.
As for the study entitled "Digitalisation for Mama Mboga: the involvement and inclusion of women in IA Agri-tech" by Angella Ndaka, It targets women farmers and players in the agri-business value chain. It is taking place mainly in the high-potential agricultural area of Kiambu in Kenya and seeks to analyse the obstacles to women's digital inclusion, examine ways of articulating women's perspectives in the development and application of agricultural AI systems, as well as in the associated policy processes.
The circular economy is a major challenge in Africa, which is why , Shamira AHMEDThe Data Economy Policy Hub is proposing a study on "Responsible AI for Gender Equality in Africa's Circular Economy", which aims to explore the use of AI to reduce food waste at different stages of South Africa's food supply chains.
In this same dynamic and in a broader sense, Rebecca Ryakitimbo researcher at @Core23lab in the Democratic Republic of Congo and members of her team are proposing research entitled "Gender Mainstreaming in AI: A Francophone and Anglophone East African Perspective" that tackles the issue of gender inclusion in AI across countries identified within the Anglophone (Tanzania, Kenya and Uganda) and Francophone (Rwanda, Burundi and DRC) East African communities. More specifically, the research will investigate barriers to the creation and use of sex-disaggregated data for AI, gender gaps and biases in AI.
Walelign Tewabe Sewunetie, academician at Debre Markos University (DMU), has presented a project entitled "Natural Language Processing (NLP), Big Data Analytics, Human Data Mining", which aims to address the challenges of the language barrier between different local communities and the problem of gender bias in machine translation datasets. The academician and his team propose to develop a machine translation prototype capable of detecting and mitigating gender bias in Ethiopia, particularly in sub-Saharan Africa.
In almost the same vein, another study by Dr. Joyce Nakatumba-Nabende from the Artificial Intelligence Laboratory at Makerere University in Uganda entitled "Understanding gender bias in the construction of artificial intelligence models in the African context" is similarly tackling gender bias and the inclusivity of artificial intelligence tools on the African continent. The results will be used to develop a framework and guidelines for mitigating gender bias in ASR systems through equity in data collection and model building.
This work is expected to develop a model governance framework to address the ethical and gender issues surrounding the data and algorithms on which RSA systems are built.
The plurality of African languages is a source of multiple challenges in Africa, which is why Natnael Tilahun, a computer scientist at AASTU University in Ethiopia, also shared his study on "Diversity and Gender Equality in the AI Ecosystem: An SLR of African Languages", which targets the under-resourcing of local African languages and proposes to examine the current state of representation of local African languages in the global AI ecosystem, in particular the under-representation of women and minorities in the AI ecosystem in Africa, which leads to a lack of diversity and inclusion.
In Kenya, for example, gender education and other interventions to ensure gender equity, equality and inclusion remain a major challenge, Juliet Chebet Moso lecturer at Dedan Kimathi University of Technology, Kenya proposes an "Artificial Intelligence-based model for the analysis of gender inequality in STEM programmes and career projection in Kenya . Specifically, the study provides an opportunity to examine female student participation in STEM courses by investigating how gender inequality affects student placement and students' perceptions of AI in STEM programmes and their eventual career placement."
It should be noted that all the teams selected under the call for projects are joining the GRAIN network de facto.
The GRAIN network aims to function as a "collaborative space" bringing together local stakeholders with the necessary expertise, including for-profit organisations, universities, governments and research, enabling self-sufficiency by building on existing capacity. Within this framework, the beneficiaries and the members of the Grain consortium continued their work on Tuesday 21 November online to create an exchange platform for a better understanding of gender equality in AI.
Not that long ago, people lived and functioned in tight communities. Every vendor knew their customers personally and could make...
This Machine Learning Glossary aims to briefly introduce the most important Machine Learning terms - both for the commercially and...
FR 
EN 
