Exploring the development of machine learning in yarn quality prediction aims to enhance digital production management level of textile enterprises gradually. Four machine learning algorithms: linear regression, grey system model, support vector machine, and neural networks are used to analyze the current research status of yarn quality prediction models. Elucidating the network structures and optimization algorithms of two types of neural networks——artificial neural networks and deep neural networks are focused on yarn quality prediction models deficiencies and proposing solutions for these networks in the yarn quality prediction field are also dissected. Further exploration of the application of deep neural networks based on sample databases in the field of yarn quality prediction is crucial to enhance prediction accuracy andaccelerate computational speed.