A community service demand identification method based on resident profile

This paper aims to address the issue of imprecision in community service demand identification by leveraging big data and user profile technologies to provide accurate and refined services for community residents. Based on user profiles, we propose a demand identification method for identifying service demands within communities, which enhances the accuracy of community service demand identification in two key ways: firstly, when training the model to identify community service demands, human observation and insight are integrated into the training of the model; and secondly, the manual labels are propagated through label propagation technology to improve the quality of model training data. The effectiveness of this method is subsequently validated using property enterprise data. The results show that observation features and label propagation technology can markedly enhance the precision of demand identification for community services. This underscores the importance of human insight and algorithms in applying big data to improve the accuracy of grassroots community services, offering some ideas and support for refining the community service system.