Name
Abstract | ||
|---|---|---|
Thanks to social media platforms, topics are diffused online. The ones which diffuse rapidly and widely are known as viral topics. Predicting the popularity of viral topics is important to online recommendation systems and marketing services. However, the task is still challenging due to the fact that the popularity of viral topics is highly dynamic and little research has been focused on how to predict the popularity accurately for viral topics. This paper first uses data collected from the largest BBS in China to find high correlations for the short-term popularity of viral topics. Based on this finding, this paper then utilizes a time series feature space to capture the behaviors of popularity of viral topics and presents a method for predicting the short-term popularity of a given viral topic by using only data of historical popularity of the topic. This paper demonstrates that our method outperforms a baseline model and one of the most sophisticated methods in the terms of MAPE (mean absolute percentage error) and RAE (relative absolute error). Furthermore, this paper gives out how long the minimum observation period is in order to predict promptly. Finally, this paper uses two cases to show the effectiveness and simplicity of our method. HighlightsThis paper finds high correlations for the short-term popularity of viral topics, so it is rational to predict the short-term popularity by using historical popularity for viral topics.This paper defines a time series feature space to capture the behaviors of popularity of viral topics and shows that it can be used to understand how the popularity of viral topics evolves over time.This paper develops a method for predicting the short-term popularity of viral topics based on time series forecasting by using only data of historical popularity of a given viral topic. The experimental results show that it performs much better than a baseline model and one of the most sophisticated methods. |
Year | DOI | Venue |
|---|---|---|
2016 | 10.1016/j.neucom.2015.10.143 | Neurocomputing |
Keywords | Field | DocType |
Popularity prediction,Viral topics,Time series forecasting,Viral topic behavior | Mean absolute percentage error,Recommender system,Data science,Time series,Feature vector,Social media,Computer science,Popularity,Artificial intelligence,Machine learning | Journal |
Volume | Issue | ISSN |
210 | C | 0925-2312 |
Citations | PageRank | References |
10 | 0.56 | 22 |
Authors | ||
5 | ||
Name | Order | Citations | PageRank |
|---|---|---|---|
| Ying Hu | 1 | 13 | 2.30 |
| Changjun Hu | 2 | 130 | 27.56 |
| Shushen Fu | 3 | 12 | 1.28 |
| Peng Shi | 4 | 20 | 5.79 |
| Bowen Ning | 5 | 10 | 0.56 |