University of São Paulo: Through Artificial Intelligence, a solution created at USP facilitates decision-making to combat hunger

An unprecedented solution that uses Artificial Intelligence could contribute to the fight against one of the biggest problems faced in Brazil today: hunger. Through the analysis and interpretation of information from hundreds of different databases, new algorithms developed by researchers from the Center for Artificial Intelligence (C4AI) at USP are able to identify which regions of a given city lack adequate infrastructure and options for the food supply. Innovation seeks to show where the public power, a company or other type of organization should act to make more accurate and efficient decisions to face food insecurity.

One of the prototypes created by the researchers was designed to identify regions most susceptible to hunger in the city of São Paulo, based on the demands of two Interinstitutional Groups that work in the municipality with a focus on actions and policies to improve food quality. In this case, the solution has been fed with open data from the federal government, the Brazilian Institute of Geography and Statistics (IBGE) and the National Supply Company (Conab), in addition to information related to the Food Industry, such as the number of establishments that they sell food by region – whether fairs, markets or cafeterias. “The integration of diverse and widespread information is still a problem in Brazil, also because of the size of the territory. There are many separate databases. In smaller and richer countries it is much easier to merge them, but we are looking to integrate more than 100 databases. Now, with our solution that cross-references information automatically, Artificial Intelligence presents the most critical locations and several other types of analysis”, emphasizes Alexandre Delbem, professor at the Institute of Mathematical and Computer Sciences (ICMC) at USP, in São Carlos, and a researcher at C4AI, which is funded by IBM and the São Paulo Research Foundation (Fapesp).

The professor, who created the new technology in just 11 days, points out that it can map multiple areas and shape itself autonomously to different types of information: “It adapts to the data it has and discovers, alone, how it can cross those it complement each other, even if they are few, and make the best use of it to contribute to decision-making. Simply put raw, untreated information that Artificial Intelligence interprets and presents results and predictions in reports, spreadsheets, graphs or maps,” he explains. As the technology can be adapted to work with any type of data, Artificial Intelligence can still gather information about climate change and predict possible shortages in agriculture, in addition to helping to overcome drought during a period of drought. “The farmer could prepare for this in advance. Each decision maker will analyze the aspects of interest. The solution could be supplied with any other type of data”, explains Delbem. Innovation can also facilitate communication between professionals from different areas, as technology is able to gather data from different areas on the same object. “It connects information and presents reliable data for decision making. Managers will be able to talk. This can help city halls by making secretaries from different areas make decisions together”, describes the scientist. Innovation can also facilitate communication between professionals from different areas, as technology is able to gather data from different areas on the same object. “It connects information and presents reliable data for decision making. Managers will be able to talk. This can help city halls by making secretaries from different areas make decisions together”, describes the scientist. Innovation can also facilitate communication between professionals from different areas, as technology is able to gather data from different areas on the same object. “It connects information and presents reliable data for decision making. Managers will be able to talk. This can help city halls by making secretaries from different areas make decisions together”, describes the scientist.

According to Professor Antonio Mauro Saraiva, from the Polytechnic School of USP and a researcher at C4AI, the case of hunger is a problem of many dimensions and, therefore, the integration of data is essential for its solution. “They are hungry for lack of food, hunger for quantity, hunger for quality, hunger for lack of access, for lack of nutrients, hunger because there was no production that year. It is a real and very serious problem that we need to solve in a concrete way. The information for each of these areas is in different banks. They are geographic, sanitary, medical data and each one of them represents a different world. The integration of this information was a huge obstacle, but one that we are solving”, observes the specialist,

In the country, the situation is serious, even more so with the increase in poverty. According to the National Survey on Food Insecurity in the context of the Covid-19 pandemic in Brazil, developed and released in 2021 by the Brazilian Research Network on Food and Nutrition Sovereignty and Security, 19 million people were without meals for a day or two. more, at the time the study was carried out. “There are a lot of hungry people, with food and health problems. The country is diverse and unequal, so it is difficult to think in general terms. If we do not have detailed information, it is very difficult to make correct decisions. We are contributing to try to find a way. This is a big challenge, but technology can help a lot. Having access to reliable data, it is possible to analyze, see what actually happens,

Created from a demand from the USP Rectory itself, through the Working Group on Public Policies to Combat Hunger and Food Insecurity, the interdisciplinary project for the creation of the new technology is the result of different partnerships. “We have researchers in the areas of social sciences, agriculture, nutrition, economics, computer engineering, among others. We are working with several institutes within USP, such as the Institute for Advanced Studies (IEA), and partnering with other universities. All relevant countries are working with this type of development. The United States and China invest heavily in Artificial Intelligence for application in social policies. In fact, in recent years, billionaire investments have been announced in large research centers. This type of study is essential”, says Saraiva.

For Professor Delbem, Artificial Intelligence can increasingly enhance the use of available resources, but for that, the path between the university, the public power and society must still be better paved: “This type of technology is increasingly increasingly present in society and with its more evident functions. The creation of C4AI encourages research in this direction, which is very important for different types of services. We developed research that led us to these techniques that prove that it is worth investing in the area, as they have great potential. We continue to promote improvements, seeking other benefits and we are open to new partnerships”, he concludes.