IFPRI CGIAR
Subscribe

Back

What we do

With research staff from more than 60 countries, and offices across the globe, IFPRI provides research-based policy solutions to sustainably reduce poverty and end hunger and malnutrition in developing countries.

benin_samuel_0

researcher spotlight

Samuel Benin

Samuel Benin is the Acting Director for Africa in the Development Strategies and Governance Unit. He conducts research on national strategies and public investment for accelerating food systems transformation in Africa and provides analytical support to the African Union’s CAADP Biennial Review.

Where we work

Back

Where we work

IFPRI currently has more than 600 employees working in over 80 countries with a wide range of local, national, and international partners.

IFPRI Regional Programs

IFPRI Country Programs

Research by country and region

Subscribe

experts

Natural Resources and Resilience

Jawoo Koo

Senior Research Fellow

Follow

Jawoo

at:

ORCiD
profile

Bio

Jawoo Koo joined IFPRI in July 2007 as a postdoctoral fellow in the Environment and Production Technology Division. He holds a B.A. in Agricultural Biology from Korea University, Seoul, Korea, and a M.S. and a Ph.D. in Agricultural and Biological Engineering from University of Florida. For his dissertation, he worked for the Soil Management CRSP, funded by USAID, in collaboration with University of Ghana and the Savannah Agriculture Research Institute, to estimate soil carbon sequestration potential in smallholders’ farming systems in Ghana using crop systems models and field surveys.

Publications

Publications

Resource thumbnail

Report

Longa: An automated speech recognition tool for Bantu languages

2023Mganga, Nelson; Jones-Garcia, Eliot; Monsalue, Andrea Gardeazabal; Koo, Jawoo
Details

Longa: An automated speech recognition tool for Bantu languages

Farm Radio International (FRI) and the CGIAR Research Initiative on Digital Innovation have col laborated on the development of an end-to-end, automatic speech recognition pipeline for the tran scription, translation, and analysis of Swahili and Luganda. This task is particularly challenging due to the number of languages used by FRI’s clients and the limited training data available for speech recognition in African languages. The tool is named ‘Longa’, or ‘Let’s chat’ in Swahili. Longa will be used to answer the surplus of phone calls currently being received from smallholder farmers asking questions about radio programs which FRI does not presently have the capacity to address. When fully implemented, Longa should allow FRI to design their broadcasts more intricately in line with the needs of farmers and better deliver insights to those most in need, such as female and youth farmers. Key results from the collaboration include a series of design principles iteratively and col laboratively developed to reflect the common values and goals of FRI and the CGIAR, a proof of concept for Longa, building on open-source models and open access corpora, to be shared with the developer community upon completion of the final tool, a 10% improvement upon the state-of-the art automatic speech recognition in Luganda radio-speech performance and accuracy, some im provement in performance with audio enhancement processes using real-world data, and proof that fine-tuning is an effective approach to expanding Longa to new languages. The next steps of the collaboration will focus on the analysis and interpretation of an aggregation of farmer phone calls and integration with the existing FRI workflow and software.

Year published

2023

Authors

Mganga, Nelson; Jones-Garcia, Eliot; Monsalue, Andrea Gardeazabal; Koo, Jawoo

Citation

Mganga, Nelson; Jones-Garcia, Eliot; Monsalue, Andrea Gardeazabal; and Koo, Jawoo. 2023. Longa: An automated speech recognition tool for Bantu languages. Digital Innovation Technical Report December 2023. Washington, DC: International Food Policy Research Institute (IFPRI).

Keywords

Africa; Artificial Intelligence; Innovation Adoption; Languages; Farmers

Language

English

Access/Licence

Open AccessCC-BY-4.0

Source

Source record

Project

Digital Innovation

Record type

Report

Resource thumbnail

Working Paper

Can we trust large language models to summarize food policy research papers and generate research briefs?

2023Kim, MinAh; Koo, Jawoo; Jung, Yunchul
Details

Can we trust large language models to summarize food policy research papers and generate research briefs?

Generative large language models (LLMs), while widely accessible and capable of simulating policy recommendations, pose challenges in the assessment of their accuracy. Users, including policy analysts and decision-makers, bear the responsibility of evaluating the outcomes from these models. A significant limitation of LLMs is their potential to overlook critical, context-specific factors. For example, in formulating food policies, it is vital to consider regional climate and environmental variables that influence water and resource availability. Nonetheless, due to their reliance on word sequencing probabilities from training datasets, LLMs might propose similar policies for distinct regions. Despite these limitations, LLMs offer considerable advantages for rapid policy analysis, particularly when resources are constrained. They serve as quick, accessible, and cost-effective tools for policy research and development, requiring minimal training and infrastructure. In our study, we assessed the efficacy of LLMs in generating policy briefs by inputting an IFPRI discussion paper into three different LLM-based approaches: a standard chatbot without extra data, a Retrieval Augmented Generation model integrating semantic search with LLM, and a custom-developed Brief Generator designed to create policy summaries from AI-analyzed paper structures. Our findings revealed that none of the LLM-generated briefs fully captured the original paper’s intent, underscoring the need for further research. Future investigations should focus on gathering more empirical data with diverse text types and volumes to better understand these outcomes.

Year published

2023

Authors

Kim, MinAh; Koo, Jawoo; Jung, Yunchul

Citation

Kim, MinAh; Koo, Jawoo; and Jung, Yunchul. 2023. Can we trust large language models to summarize food policy research papers and generate research briefs? Digital Innovation Working Paper December 2023. Washington, DC: International Food Policy Research Institute (IFPRI).

Keywords

Policies; Artificial Intelligence; Technological Changes; Data

Language

English

Access/Licence

Open AccessCC-BY-4.0

Source

Source record

Project

Digital Innovation

Record type

Working Paper

Resource thumbnail

Blog Post

Creating a Virtual Food Policy Expert Using Artificial Intelligence: Advantages and Disadvantages of Customized Chatbot

2023Koo, Jawoo
Details

Creating a Virtual Food Policy Expert Using Artificial Intelligence: Advantages and Disadvantages of Customized Chatbot

Artificial intelligence (AI) tools are rapidly transforming the way people work, providing opportunities for improving efficiency, accelerating innovations and supporting human decision-making. The area of food policy research and policymaking is no exception. In a previous blog, we argued that policymakers should not rely exclusively on AI-powered chatbots (e.g., Bard, ChatGPT, and Claude), given that the outputs of current versions of AI simply draw upon a vast number of existing passages in publications written by humans, assembled without logic and, consequently, prone to errors. In addition, many chatbots do not identify the sources used for assembling specific policy information, thus potentially undermining trust and credibility. Still, the potential of having a virtual assistant at your fingertips is so appealing that we wanted to test it ourselves.

Year published

2023

Authors

Koo, Jawoo

Citation

Koo, Jawoo. 2023. Creating a virtual food policy expert using artificial intelligence: Advantages and disadvantages of customized chatbot. Agrilinks. First posted on December 21, 2023. https://agrilinks.org/post/creating-virtual-food-policy-expert-using-artificial-intelligence-advantages-and-disadvantages

Keywords

Artificial Intelligence; Food Policies; Expert Systems; Knowledge-based Systems

Language

English

Access/Licence

Limited Access

Source

Source record

Project

Digital Innovation

Record type

Blog Post

Blogs

Can we trust AI to generate agricultural extension advisories?
IFPRI Blog
 | 
December 05, 2023

Can we trust AI to generate agricultural extension advisories?

By Jawoo Koo, Medha Devare, and Brian King

A chatbot experiment.

How can artificial intelligence-powered chatbots help policymakers? A roadmap for Kenya
IFPRI Blog
 | 
June 01, 2023

How can artificial intelligence-powered chatbots help policymakers? A roadmap for Kenya

By Boniface Akuku, Clemens Breisinger, Joseph Karugia, Jawoo Koo, Michael Keenan, and Richard Ndegwa

Integrating AI into the policy process.

Rural populations face heightened COVID-19 risks
IFPRI Blog
 | 
May 12, 2021

Rural populations face heightened COVID-19 risks

By Jawoo Koo, Claudia Ringler, Aniruddha Ghosh, and Carlo Azzarri

An analysis of 12 low- and middle-income countries shows potential pandemic hotspots.

Events

ICTforAg 2024: Localizing impact
IFPRI at External Event

 | 

May 28, 2024, 9:00 – 5:00 pm

ICTforAg 2024: Localizing impact

Event: May 28, 2024 – 9:00am to 5:00pm EDT. This event explores the localization of ICT, responsible AI in agriculture, the potential for AI to increase farm profitability, empowering citizen science, and harnessing digital solutions for sustainable and regenerative farming practices.

ICTforAg 2023: Cultivating Inclusion
IFPRI at External Event

 | 

November 7 to 9, 2023

ICTforAg 2023: Cultivating Inclusion

Virtual Conference: November 7, 2023 – 06:30 AM to November 9, 2023 – 5:00 PM EST. IFPRI is participating in ICTforAg 2023. During this 3-day virtual event, agricultural stakeholders and technology experts will come together to share knowledge, find solutions, and form partnerships to address challenges in agri-food systems across low- and middle-income countries.

IFPRI @ Adaptation Futures 2023
IFPRI at External Event

 | 

October 2 to 6, 2023

IFPRI @ Adaptation Futures 2023

Event: October 2, 2023 – 9:00 AM to October 6, 2023 – 6:00 PM EDT. IFPRI is participating in Adaptation Futures 2023 conference. The AF2023 Conference seeks to learn from Indigenous, local knowledge and voices in climate change adaptation research, policies, practices, and actions around the world.

News

September 24, 2020 | In the News

Hewlett Packard Enterprise Advances the Global Food System Through Memory-Driven Computing With CGIAR (Digital Journal) 

Digital Journal republished a press release from Hewlett Packard announcing a partnership with the CGIAR. Researchers need to generate a timely, high-frequency picture of what is happening in “food basket” locations – or areas of significant food production – around the world. A complete picture often requires data from multiple sources including crop performance, weather records, economic activity, […]

September 12, 2020 | In the News

ECONOMÍA. Saocom: satellite data fill the void of dwindling crop tours (El Intransigente)

El Intransigente (Argentina) published an article stating that the pandemic is helping to usher in a new era of food-production forecasts that rely more on satellite data and artificial intelligence and less on information gathered by people. Rather than sending people to visit things, the market is looking for remote sensing. Still, there are shortcomings in some satellite […]

June 15, 2018 | In the News

Is There an Algorithm to End World Hunger? (Food Tank)

Food Tank published a report recognizing the use of Big Data and innovative technology as “the next agricultural revolution.”