The Guangdong, Hong Kong and Macao Greater Bay Area constructed by China, as a newcomer of the inter-fourth bay area, its current development is focused on infrastructure and economic development, and its construction in branding is weak. In this paper, crawler software is used to crawl the data, and feature selection and weighting are performed based on text data through TF-IDF weighting method. The text feature extraction method based on statistics is used to mine the brand personality vocabulary of Guangdong, Hong Kong and Macao Bay Area, and determine the prominent brand personality of Guangdong, Hong Kong and Macao Bay Area. A semantic network is constructed and EDT-based text sentiment analysis is used to obtain tourists’ perception of the image of the Guangdong-Hong Kong-Macao Greater Bay Area. The results show that tourists’ positive emotions accounted for 78.98%, neutral emotions accounted for 10.24%, and negative emotions accounted for 10.78%, of which positive and neutral emotions accounted for 89.22%. Combined with the analysis results, the digitalization strategy for brand image of Guangdong, Hong Kong and Macao Greater Bay Area is proposed.