The development of painting art industry has raised higher demands for the property value assessment of paintings. To address the challenges in evaluating the property value of paintings, we first extracted the value-related factors using digital intelligence technology. After expert validation, we constructed a four-dimension index system of “Social - Economic - Technological - Ecological” Benefits. And then, we developed a property assessment model for paintings’ property value, specifically the LSTM-GRU-Attention model based on the index system. Experiments on the value assessment of paintings were conducted with real transaction data collected from the Artron website to examine the established index system and value assessment model. The results indicate that the new model which integrates data intelligence technologies shows an accuracy of nearly 80%, i.e. a scientific assessment of paintings’ value. The approach also offers valuable insights and references for the development of other art industries.