Arizona State University, in collaboration with the Jane Goodall Institute Global and researchers from ASU’s Institute of Human Origins, is leveraging artificial intelligence (AI) and machine learning to digitize over 60 years of handwritten chimpanzee research from Tanzania’s Gombe National Park. This initiative aims to convert thousands of paper-based “Tiki sheets” and related field materials into structured, searchable data that can be analyzed on a large scale.
The project involves ASU primatologist Ian Gilby, student researcher Joesh Jhaj, and senior AI development engineer Krishna Sriharsha Gundu from ASU Enterprise Technology’s AI Acceleration team. By digitizing these extensive records, the team seeks to continue the legacy of renowned scientist and conservationist Jane Goodall, while advancing research efforts in understanding chimpanzee behavior and ecology.
The Gombe archive, which has been hosted at ASU since 2022, consists of daily observations of wild chimpanzees recorded over six decades. Researchers in Tanzania tracked a single “focal” chimpanzee each day, documenting data such as arrival and departure times, feeding behavior, and interspecies encounters on standardized sheets. As a result, hundreds of thousands of records have accumulated, making manual data entry a time-intensive task for student researchers.
“There’s a lot of value to these data,” Gilby stated, emphasizing their importance in providing insights into human origins and the complex nature of chimpanzee behavior. “A better understanding of their biology and behavior gives a better chance of protecting this iconic endangered species,” he added. To expedite the digitization process, Gilby collaborated with ASU’s AI Acceleration team, leading to the development of a novel solution that combines computer vision techniques with machine learning.
Gundu’s innovative approach employs imaging software to straighten scanned documents and extract structured data points, converting the information into spreadsheet format that can be incorporated into a relational database for further analysis. “We combined this traditional AI technology with newer large language models to review and analyze the handwritten notes written in the margins on the sheets,” Gundu explained.
This project utilizes computer vision for layout detection and data extraction, while generative AI is tasked with interpreting handwriting. Jhaj elaborated, stating, “Computer vision translates an image that the computer can then pick up on,” while generative AI aids in reading and translating the handwriting on the documents. Building on Gundu’s preprocessing efforts, Jhaj developed specialized code for translating contextual marks and handwritten notes into structured research data, employing GPT’s API when clarity is needed for ambiguous handwriting.
In the digitization process, students working with the archive compare outputs against the original handwritten sheets to ensure accuracy. The team anticipates that the Gombe AI tool will significantly reduce manual entry time, enhance analytical consistency, and better integrate Tiki sheet data with other materials, including handwritten protocols, video, and geospatial datasets, that are part of the broader Gombe AI Research Platform.
The project serves as a noteworthy example of AI’s application in academic research, particularly in accelerating archival digitization and enabling large-scale analysis of legacy datasets. For universities and research institutions, it underscores how AI tools can enhance the utility and longevity of historical data collections. In fields like anthropology, ecology, and conservation science, decades of handwritten material often remain underutilized due to the challenges of digitization.
By integrating computer vision and generative AI into research workflows, ASU is testing a scalable model that may be replicated by other institutions grappling with paper-based archives. As the adoption of AI technologies in research accelerates, initiatives like the Gombe AI project signal a significant shift from experimental applications to operational tools that are becoming embedded in research infrastructures.
See also
AI Study Reveals Generated Faces Indistinguishable from Real Photos, Erodes Trust in Visual Media
Gen AI Revolutionizes Market Research, Transforming $140B Industry Dynamics
Researchers Unlock Light-Based AI Operations for Significant Energy Efficiency Gains
Tempus AI Reports $334M Earnings Surge, Unveils Lymphoma Research Partnership
Iaroslav Argunov Reveals Big Data Methodology Boosting Construction Profits by Billions
