Polymers occupy an important place in modern society and are widely used in daily life and industrial production. However, the complexity of polymer chemistry and morphology poses challenges for their advancement. Integrating machine learning algorithms and extensive data resources has paved a new path for advancements in polymer science and engineering through data-driven methods. Here, we present the application of QSPR methods to polymer properties, including model development, property prediction, and molecular design. In this section, we summarize the steps of machine learning (ML) model building and the methodology, review the work on ML-based implementation of polymer property predictionPolymer property prediction, and examine the application of ML methods in polymer design.