This paper utilizes text mining methods in the field of big data mining to perform operations such as mining and natural language processing on network texts related to farming culture IP and rural branding. Based on the results of the network text data mining, the image design of farming culture IP and rural branding are explored in the integrated media environment. Audience perception of rural IP image and advertising strategy of rural brand are explored through data mining methods. The order of audience attention to the rural IP image of traditional farming villages in Zhejiang Province from highest to lowest is: rural resources (38.24%), rural atmosphere (29.12%), rural environment (27.82%), and rural activities (4.82%). Rural brand advertising in Zhejiang Province mainly focuses on 3 categories: farming landscape, agricultural products and rural tourism. In terms of media type, it favors mobile placement, with Non OTV and outdoor advertising as the main ones. The advertising forms with the highest placement frequency are full-screen ads (5.47%) and infomercials (5.51%).