Abstract | ||
---|---|---|
Recently, with rapid growth of news media in the Internet, it presents both challenges and opportunities. One challenge lies in how to automatically extract a small group of company-specific keyphrases from news media that can accurately describe a company. Company-specific keyphrase extraction is an ef cient way to mine information from the news article. There are mainly two kinds of approaches for keyphrase extraction: supervised and the unsupervised. In this paper, we propose entity-rank, a novel unsupervised model which is based PageRank and integrate it with the specific company entity information. The experiment result shows that our model has an improvement compared with several other baseline models. |
Year | Venue | Field |
---|---|---|
2017 | CIS | PageRank,Microsoft Windows,Information retrieval,Computer science,News media,Feature extraction,Artificial intelligence,Machine learning,Semantics,The Internet |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
0 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Qiang Pan | 1 | 0 | 0.68 |
Ping Guo | 2 | 601 | 85.05 |
Xin Xin | 3 | 58 | 7.73 |
Junshuai Liu | 4 | 0 | 0.34 |