抄録
Offer Organization: Japan Society for the Promotion of Science, System Name: Grants-in-Aid for Scientific Research, Category: Grant-in-Aid for Scientific Research (C), Fund Type: -, Overall Grant Amount: - (direct: 3900000, indirect: 1170000)
Disclosure of personal information in social media is serious social problem. We developed a technology that detects revelation of personal information from natural language text to be posted in social medial. Evaluation using 1000x11 texts posted by 11 volunteers has shown that our developed technology detected 85% of revelations of volunteers' affiliations and job categories. On the other hand, privacy infringement by linking multiple fragments of personal information become a social concern.We developed a technology that detects texts in social media posted by a person of interest by linking the texts with a resume of the person. We have evaluated our developed technology using 1000x12 texts posted by 12 volunteers and 1000x100 texts posted by 100 background persons. The evaluation has shown that our developed technology worked successfully for 8 volunteers, i.e. it identified texts posted by the volunteer from texts posted this volunteer and 100 other persons.