Hames Sherif.

On a mission to support the global campaign for zero hunger, a graduate from Rochester Institute of Technology (RIT) Dubai has published a book to share her findings on how to predict staple crop yields in arid and desert climates.

Using a branch of artificial intelligence known as machine learning, Hames Sherif has applied her graduate research studies to develop a model that could help farmers to manage their resources more efficiently and aid long-term food security.

She inspired by the United Nation Sustainable Development Goal 2, which aims to create a world free of hunger by 2030, Hames set out to explore how she could use her passion for data and statistics to address food security.

Through a study conducted as part of her Master of Science in Professional Studies: Data Analytics degree, Hames used a common algorithm to create a tool that can estimate more accurately how much crop will be produced in a season.

The resulting research paper attracted attention from the international academic community and has subsequently been published in a book by Eliva Press. Revealing the motivation behind her work Hames said, “I discovered that in 2009, the Food and Agriculture Organization of the United Nations predicted that agricultural production would need to increase by 60% to sustain a population that is expected to exceed 9 billion by 2050. An important part of the challenge for policy makers is predicting crop production. Traditionally farmers depended on their expertise to estimate crop yields and make decisions accordingly, however, data mining techniques can offer a more reliable approach.”

Originally from Egypt, Hames added, “For a long time, African nations have been experiencing food insecurity, and so the research is centered on those countries in Africa that have desert and semi-arid climates. I decided to focus on the staple crops like maize, wheat and rice, which are the essential produce needed to sustain a population, and cover their basic nutritional needs. As part of my research, I also explored the factors that affect the fluctuations in crop yields.”

Explaining the results of her research, Hames said, “I tested various models to arrive at a solution that relies on domain expertise as well as historical data to build a prediction model. Interestingly, the widely known factors like fertilizers, precipitation and temperature, were not amongst those to have the highest effect on crop yield. What’s more, factors like the uptake of agricultural machinery and population growth have beneficial effects on staple crop yields, as more specialized physical and human resources lead to the adoption of more advanced farming practices.”

Speaking about Hames’ research and her contribution to the field, Dr. Khalil Al Hussaeni, Assistant Professor of Computing at RIT Dubai, said, “When Hames approached me for supervision, I was very intrigued by this topic for being a humanitarian cause. One of the major challenges we had in this endeavour was the lack of detailed and comprehensive data, on which the entire study would be based. I am happy to see that, with strong motivation and will power, Hames was able to overcome those challenges."

"We hope that the results of this study would be a vital contributors to the advancement of global food security.” Sharing her own aspirations for the future of the project, Hames concluded, “I’m grateful that my research has been published and become available to the international community. I hope that it will encourage more investment in crop yield data collection, which would greatly improve the forecasts and help to make an even bigger impact on global food security.”