Our nation's economic foundation is agriculture, which is also the general term for the production of human food. Turmeric plant production materials play a role in our daily lives, such as food preparation and health care. The most common turmeric plant diseases are Leaf Spot and Leaf Blotch, which are airborne diseases that can wipe out entire crops in a matter of days. In this paper, Hybrid algorithms such as Regression Tree, Kernel least square regression, Quadratic SVM, Ensemble boosted tree, optimized ensemble, square exponential GPRand Optimized neural networkare proposed to predict the growth of turmeric and turmeric plant disease. The proposed hybrid algorithms are predicting the turmeric growth level from turmeric yield dataset. It is also used in the agricultural sector to increase crop yield by spotting disease affliction early and categorising the type of disease affected so that preventative measures can be taken to stop the disease from spreading to other plants in the field.Regression Tree, Kernel least square regression, Quadratic SVM, Ensemble boosted tree, optimized ensemble and square exponential GPRalgorithms are performing less due to prediction of turmeric growth based on accuracy and time. These algorithms give huge difference in prediction such as accuracy level and speed. Turmeric yield data fed to the pre-trained for prediction of turmeric growth. The proposed Hybrid algorithm gives high accuracy prediction compared to other statistical algorithms. Optimized neural networkgives high accuracy of about 96% compared to conventional methods.