Construction of optimal environment model for growth period of Phalaenopsis aphrodite based on semi supervised learning combined with least squares support vector machine

As an important ornamental plant, growth environment of Phalaenopsis aphrodite has a significant impact on its growth and development．Traditional cultivation methods rely heavily on experience and lack scientificity and precision．Environmental parameters and growth status data during growth process of Phalaenopsis aphrodite were collected, key features were screened, and semi supervised learning combined with least squares support vector machine was used to train a deep learning model to predict optimal environmental conditions for Phalaenopsis aphrodite growth．Through self-learning methods, model could select high confidence samples from a large number of unlabeled samples, increase number of training samples, and improve model's generalization ability and prediction accuracy．Experimental results show that when probability threshold was set to 97%, model accuracy was the highest, with root mean square error of 3．974 and coefficient of determination of 0．975．This model can provide a new solution for the scientific cultivation of Phalaenopsis aphrodite．