This study aims to explore users' perceptions, emotions and social network relationships, as well as the differences among them, regarding the image of major public cultural institutions (museums, libraries and galleries) in mainland China. The findings provide essential evidence for improving the image of libraries on new social media and enhancing service quality.
This study employs an integrated approach that integrates quantitative and qualitative research methods. Based on the big–data analysis and mining tool ROST CM6, this study utilizes word-frequency analysis, sentiment analysis and social and semantic network analyses to examine user comments from major public cultural service institutions in mainland China posted on a new tourism social media platform (Dianping). It also conducts both an overall analysis and a differential analysis of their image.
The findings indicate that public perception of museums, galleries and libraries primarily revolves around three dimensions: functionality, environment and service. Museums and galleries are predominantly perceived as venues for visiting and appreciating exhibited artifacts or artworks, while libraries are mainly associated with book borrowing and reading. The distribution of emotional tones in user reviews follows a similar pattern: with positive comments predominating, with moderate and negative feedback being relatively scarce. These three public cultural institutions form a social and semantic structure characterized by a “core layer–secondary core layer-periphery layer” hierarchy.
Based on the above analysis, this paper proposes three suggestions for enhancing the image of public cultural service institutions. Specifically, in terms of perception, it is necessary to expand the tourism functions of public cultural institutions and explore their cultural connotations; in terms of emotion, it is important to reduce users' negative emotions and strengthen their positive emotions; and in terms of social and semantic networks, attention should be paid to the peripheral levels and long-tail nodes.
