Name
Abstract | ||
|---|---|---|
With the development of information technology, there is explosive growth in the number of online comment concerning news, blogs and so on. Good comments can improve the experience of reading, but the massive comments are overloaded, and the qualities of them vary greatly. Therefore, it is necessary to predict popular comments from all the comments. In this work, we introduce a novel task: popular comment prediction (PCP), which aims to find out which comments will be popular automatically. First, we construct a news comment corpus: Toutiao Comment Dataset, which consists of news, comments, and the corresponding label. Second, we analyze the dataset and find the popularity of comments can be measured in three aspects: informativeness, consistency, and novelty. Finally, we propose a novel multi-target text matching model, which can measure these three aspects by referring to the news and surrounding comments. Experimental results show that our method can outperform various baselines by a large margin on the new dataset. |
Year | DOI | Venue |
|---|---|---|
2019 | 10.1007/978-3-030-32233-5_48 | Lecture Notes in Artificial Intelligence |
Keywords | DocType | Volume |
Application,News comment,Deep learning | Conference | 11838 |
ISSN | Citations | PageRank |
0302-9743 | 0 | 0.34 |
References | Authors | |
0 | 4 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Deli Chen | 1 | 7 | 2.81 |
| Shuming Ma | 2 | 83 | 15.92 |
| Pengcheng Yang | 3 | 7 | 5.15 |
| Qi Su | 4 | 69 | 13.73 |