Recipe popularity prediction based on the analysis of social reviews

In social based Web services systems, some resources gain popularity while others do not. It would be valuable if we can predict the popularity of certain resource. In this work, we study the recipe popularity prediction problem using the Yelp dataset. We investigate various features that can be extracted and help to improve the performance. In particular, we propose to do the sentiment analysis over the reviews and treat the sentimental scores as one of the features. A polynomial regression model is developed to predict the recipe popularity. The experimental results show that our proposed method outperforms the baseline method.