Blog screening and mining based on temporal features and hybrid search in big data
Concerning that the correlation degree of the existing methods of blog screen and mining is loose and the information retrieval of the methods is deficient,a method based on temporal feature and hybrid search method was proposed.Considering the user reviews are important sources of evidence combinat...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Beijing Xintong Media Co., Ltd
2017-01-01
|
| Series: | Dianxin kexue |
| Subjects: | |
| Online Access: | http://www.telecomsci.com/zh/article/doi/10.11959/j.issn.1000-0801.2017001/ |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Concerning that the correlation degree of the existing methods of blog screen and mining is loose and the information retrieval of the methods is deficient,a method based on temporal feature and hybrid search method was proposed.Considering the user reviews are important sources of evidence combination,the average number of reviews for blogs,the sources of BM25 relevance scores,the longest blog BM25 scores and time range between the latest related blog paper and the oldest related blog paper are being as the temporal feature sets.In addition,considering local search advantage of linear search(LS) and global search advantage of differential evolution(DE),the two kinds of information search methods were combined.BlogS06 data set was used in the experiment which was consists of blog home pages,XML source files and its blog portal pages,it was used for TREC 2007 and TREC 2008 blog mining experiments.Experimental results show that the proposed method can obtain satisfactory results in terms of running time and effectiveness. |
|---|---|
| ISSN: | 1000-0801 |