Name
Abstract | ||
|---|---|---|
We have developed a system for identifying the person who posted posts of interest. It calculates the similarity between the posts of interest and the resume of each candidate person and then identifies the resume with the highest similarity as that of the posting person. Identification accuracy was improved by using the posts of persons other than the target person. Evaluation using 30 student volunteers who permitted the use of their resumes and sets of tweets showed that using information from tweets of other persons dramatically improved identification accuracy. Identification accuracy was 0.36 and 0.53 when the number of other persons was 4 and 9, respectively. Those that the target person can be limited in 10 % of the candidates were 0.72 both with 4 and 9 such employees. |
Year | DOI | Venue |
|---|---|---|
2015 | 10.1007/978-3-319-26762-3_22 | Lecture Notes in Business Information Processing |
Keywords | Field | DocType |
Social network,Privacy,Security,Anonymity | Internet privacy,Social network,Computer science,Knowledge management,Anonymity | Conference |
Volume | ISSN | Citations |
228 | 1865-1348 | 0 |
PageRank | References | Authors |
0.34 | 11 | 5 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Yohei Ogawa | 1 | 0 | 0.34 |
| Eina Hashimoto | 2 | 0 | 0.34 |
| Masatsugu Ichino | 3 | 24 | 7.61 |
| Isao Echizen | 4 | 299 | 68.82 |
| Hiroshi Yoshiura | 5 | 3 | 2.84 |