AUC of prediction results using different adopter features within t early hours.

† A linear combination with coefficients determined by regression fitting using least squared error.Prediction features include the number of followers (fol), the number of tweets (twt), the diversity of topical interests of adopters (H1), and the number of early adopters (n). The threshold is expressed as a top percentile of most popular hashtags that are deemed viral for evaluation purposes. Best results for each column are bolded.