In order to better complete the evaluation of used sailboat prices, this article succeeded in having a system that can predict the prices of used sailboats and combined various analytical methods to realize the prediction of Hong Kong sailboat prices. This article focuses on the following issues: identifying mathematical models for rational pricing, explaining the effect of regional variables on prices, and simulating the regional impact of sailboat prices. This article consists mainly of the following: First, neural networks and deep learning related models are developed to measure the impact of sailboat characteristics on their pricing from multiple perspectives, and predictions are made for the used sailboats based on these models. Second, a variety of geographical factors are taken into account to analyze the correlation of regions, and regional variables with strong correlation analysis are added and combined with BiLSTM-AT model to explain the influence of regions on listing prices. Then, data on the corresponding variables in Hong Kong were collected, and cluster analysis was performed on regionally relevant factors to construct a multi-regional cluster price model. Last, sensitivity analysis and robustness analysis are performed on the completed model.