AUTHOR=Mokhtar Ali , El-Ssawy Wessam , He Hongming , Al-Anasari Nadhir , Sammen Saad Sh. , Gyasi-Agyei Yeboah , Abuarab Mohamed TITLE=Using Machine Learning Models to Predict Hydroponically Grown Lettuce Yield JOURNAL=Frontiers in Plant Science VOLUME=Volume 13 - 2022 YEAR=2022 URL=https://www.frontiersin.org/journals/plant-science/articles/10.3389/fpls.2022.706042 DOI=10.3389/fpls.2022.706042 ISSN=1664-462X ABSTRACT=Prediction of crop yield is an essential task for global food supply, particularly in the developing countries. This study investigated lettuce yield (fresh weight) prediction using four machine learning (ML) models; Support vector regressor (SVR), extreme gradient boost (XGB), random forest (RF) and deep neural network (DNN). It was cultivated in three hydroponics systems (suspended NFT system, pyramidal aeroponic system and tower aeroponic system) interacted with three different magnetic unit strengths under controlled greenhouse environment during two seasons in 2018 and 2019. Three scenarios consisting of combinations of input variables (leaf number, water consumption, dry weight, stem length, stem diameter) were assessed. XGB model with scenario 3 (all input variables) yielded the lowest RMSE as 8.88 g followed by SVR with scenario 3 as 9.55 g, and the highest was by RF with scenario 1 as 12.89 g. All model-scenarios, having SI index values less than 0.1 that were classified as excellent in predicting fresh lettuce yield. Based on all performance statistics, the two best models were SVR with scenario 3 and DNN with scenario 2 (leaf number, water consumption, dry weight). However, DNN with scenario 2 requiring less input variables is preferred. The potential of DNN model to predict fresh lettuce yield is promising and it can be applied on a large scale as a rapid tool for decision makers to manage crop yield.