AHMoSe: A knowledge-based visual support system for selecting regression machine learning models

Decision support systems have become increasingly popular in the domain of agriculture. With development automated machine learning, agricultural experts are now able to train, evaluate and make predictions using cutting edge learning (ML) models without need for much ML knowledge. Although this approach has led successful results many scenarios, certain cases (e.g., when few labeled datasets available) choosing among different with similar performance metrics is a difficult task. Furthermore, these do not commonly allow users incorporate their knowledge that could facilitate task model selection, gain insight into prediction system eventual decision making. To address issues, paper we present AHMoSe, visual allows better understand, diagnose compare regression models, primarily by enriching model-agnostic explanations validate describe use case scenario viticulture domain, grape quality prediction, where enables select perform better. We also discuss feedback concerning design tool from both experts.