Skip to content

InsectAI COST Action member Dr. Ferhat Sadak, presented his work at the International Conference on Machine Learning Technologies

Dr. Ferhat Sadak

At the 10th International Conference on Machine Learning Technologies (ICMLT 2025) in Helsinki, I presented my work titled “Addressing Class Imbalance in Insect Population Monitoring: A Machine Learning-Based Approach.” This work stems from a Short-Term Scientific Mission (STSM) conducted at Naturalis Biodiversity Center in collaboration with Dr. Chantal Huijbers and Dr. Laurens Hogeweg.

My presentation received strong interest from attendees with machine learning (ML) backgrounds, particularly on the topic of applying methods such as SMOTEENN and SMOTETomek to ecological datasets. Although most participants did not have a background in biodiversity or ecology, the session

triggered insightful discussions around the integration of ML with environmental challenges. I introduced the InsectAI COST Action (CA22129)—a European network that aims to advance insect biodiversity research using AI techniques—explaining its goals, working group structures, and opportunities for collaboration, such as STSMs. I encouraged attendees to visit the project website and explore how their ML expertise could contribute to this initiative.

Dr. Chantal Huijbers
Dr. Laurens Hogeweg

In addition to presenting my research, I had the honour of chairing the session on “Digital Image Analysis and Processing Methods”, where I served as Session Chair. This session featured a range of talks on cutting-edge image analysis techniques, which further aligned with my research interests and provided a valuable platform to network with fellow experts.

Throughout the conference, I engaged with numerous early-career researchers and PhD students who expressed a strong interest in transferring their ML skills to novel domains. I shared insights about the InsectAI project, offering guidance on how to get involved. Some of these researchers are now considering submitting STSM proposals under the InsectAI umbrella.

In summary, my participation in ICMLT 2025 helped promote the InsectAI COST Action to the broader machine learning community, facilitated potential new collaborations, and enhanced my own research through exposure to state-of-the-art ML methods. Chairing a session and presenting my work not only elevated the visibility of my contributions but also emphasized the growing relevance of cross-disciplinary research in addressing ecological and environmental challenges.