This study aims to classify folk paper-cut patterns by regional culture, leveraging Semantic Web and LSTM technologies to discern how these patterns reflect distinct cultural characteristics. By developing an LSTM model capable of recognizing and categorizing these patterns, our study not only demonstrates high accuracy in classifying regional cultural genes but also reveals the depth of cultural heritage embedded in paper-cut art. The findings underscore the potential of computational methods in understanding and preserving the rich tapestry of cultural expressions through paper cuts. This work sets a foundation for future explorations into the digital preservation of cultural heritage, highlighting the critical role of technology in safeguarding and interpreting traditional arts in the context of regional culture.