|
KMID : 1155220190440020181
|
|
Journal of the Korean Society of Health Information and Health Statistics
2019 Volume.44 No. 2 p.181 ~ p.188
|
|
|
Text Mining for Identifying Topics in Internet Q&A about Adolescents’ Sexual Concerns
|
|
Yu Jung-Ok
|
|
|
Abstract
|
|
|
Objectives: Because of its accessibility and anonymity, the internet is common way for adolescents to find information about sensitive issues, such as sexual concern. The purpose of this study was to explore the concerns of adolescents about sex through internet Q&A on a Korean popular portals site by applying a text mining.
Methods: The web page of the NAVER Q&A was crawled by searching the keyword using R. A total of 3,150 questions (male 1,440, female 1,710) from 2003 to 2017 on adolescent sexual concerns were used as object. Latent Dirichlet Allocation (LDA) was applied to perform text mining on the dataset.
Results: In the LDA, 7 topics were selected for male adolescents and 11 topics for female adolescents such as same sex attraction, sexual identity confusion, masturbation, genital anatomy and sexual assault.
Conclusions: This study provides useful information on sexual concerns of Korean adolescents. Topic modeling could be used as a tool to understand the online big data by discovering relevant topic.
|
|
|
KEYWORD
|
|
|
Text mining, Topic modeling, Adolescent, Sexual health, Internet
|
|
|
FullTexts / Linksout information
|
|
|
|
|
|
Listed journal information
|
|
|
|