Abstract
The Internet is deeply integrated with the traditional catering industry. The massive consumption data generated by the group-buying website of catering review implies the spatial and temporal distribution characteristics of the overall urban consumption and the level of urban catering. Based on the catering consumption data of Mianyang on Meituan website, we used Word2Vec+LSTM and naive bayes to extract user comment data for emotional analysis, and constructed multi-dimensional scoring of restaurants. We applied TextRank and TF-IDF to extract keywords and constructed restaurant labels. Combining restaurant location and visual analysis technology, we built the interactive visual analysis system, Visual Analysis for Urban Catering Data (VAUCD), that supports mining the type, location, and time preference of catering consumption. VAUCD can help understand the distribution of restaurants and the level of urban catering, and provide personalized recommendations.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have