Semantic Feedback and Identity-Responsive Design in Cultural Platforms: A Case Study of the Shu Brocade Repository

Abstract

This study focuses on the Digital Collection Repository on the official website of the Chengdu Shu Brocade and Embroidery Museum. Based on the Shu brocade pattern images displayed, to explore how User Perception, User Satisfaction and User Identity affect their Participation Intention and Feedback Behaviour. A Mixed-Methods Design was used to collect 209 Structured questionnaires and Open-ended Feedback texts. A user-platform Interaction Predictive Model was constructed. The results showed that users generally accepted the Intelligent Functions in the platform, such as AI-based Pattern Recognition and Data Clustering, However, they were dissatisfied with the Image Quality and Cultural Interpretation Depth. General Users were more positive, while Digital Experts were significantly less willing to participate. Binary Logistic Regression analysis showed that Satisfaction was a Significant Positive Predictor. Expert Identity was a Negative Moderator variable. Six Major Feedback Themes were extracted through Thematic Analysis conducted by ATLAS.ti. It covers Personalisation, Platform Transparency and Collaborative Mechanisms. Combined with Grad-CAM heat map visualisation analysis, the study further revealed the Perception Gap between Model Attention Regions and User Expectations. It also reflects the issues of Structured Visibility and Semantic Justice. Based on the existing resources, a Dual-Path Feedback Model is proposed, and an optimisation strategy is constructed accordingly. It includes Identity-Responsive Interfaces, Feedback Loops and Semantic Co-creation. To support the implementation of Inclusive Governance of Algorithmic Cultural Platforms and Cognitive Data Justice.