A Method of Analyzing Sentiment Polarity of Multilingual Social Media: A Case of Korean-Chinese Languages

Cui, Meina; Ye, Jin; Kwon, Ohbyung

doi:10.13088/jiis.2016.22.3.091

Cited by 2 publications

(2 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the KoALA program, which is suitable for the Korean language, was used, 79 conversations included emoticons or URL video links and, thus, were excluded from text mining. As emoticons enrich the information provided on SNSs and strengthen social connections, further research to analyze emoticons is required, which will enhance usability [ 50 - 52 ]. Lastly, participation patterns in chats differed by group, and although positive effects through the SNS may occur, the potential for negative consequences (eg, loneliness and social comparison) cannot be excluded [ 47 ].…”

Section: Discussionmentioning

confidence: 99%

Network Support Using Social Networking Services to Increase Exercise Adherence Among Korean-Chinese Middle-Aged Migrant Women: Mixed Methods Study

Lee

Kim

et al. 2020

JMIR Mhealth Uhealth

View full text Add to dashboard Cite

Background Social networking services (SNSs) are recognized to be a promising approach to easily deliver health interventions and to enhance social support for exercise adherence. However, the patterns and aspects of social support through SNSs have not been reported and their influence on other social-cognitive factors remains inconclusive. Objective Our objective is to explore how social support delivered through SNSs impacts interactions among Korean-Chinese (KC) middle-aged women and to identify how this approach influences social-cognitive factors for exercise (eg, sense of community, self-efficacy for exercise, and social support for exercise). Methods A mixed methods design was used. Text analysis of SNS messages and text mining using the Korean Natural Language Application (KoALA) were conducted. Social-cognitive factors (eg, sense of community, self-efficacy for exercise, and social support for exercise) were assessed at baseline and after 12 weeks using a structured questionnaire. A comparison of social-cognitive factors at baseline and at 12 weeks was conducted to identify any potential significant changes, using the Wilcoxon signed-rank test. Results A total of 259 SNS messages were collected from 24 KC women, distributed among four chat groups, who participated in a 12-week walking intervention program between August and October 2018. The individual average frequency of chatting via the SNS was 10.79 (range 0-34) and the most frequent type of social support through the SNS was network support (172/259, 66.4%). The most common words extracted from the SNS were Health, Exercise, Participation, and We. Overall, the perceived levels of sense of community (P<.001) and social support for exercise (P=.002) were significantly increased at 12 weeks compared with baseline. Group 1 (P=.03) and Group 4 (P=.03), whose members demonstrated the highest frequency of network support, experienced a significant increase only in the level of sense of community. Conclusions By integrating these data and conducting a mixed methods analysis, we observed that among the types of social support, network support was a key point for the promotion of social-cognitive factors in increasing exercise adherence. Therefore, network support through SNS-based interventions should be considered as a useful strategy to help vulnerable migrant populations make changes to exercise behaviors.

show abstract

Section: Discussionmentioning

confidence: 99%

Network Support Using Social Networking Services to Increase Exercise Adherence Among Korean-Chinese Middle-Aged Migrant Women: Mixed Methods Study

Lee

Kim

et al. 2020

JMIR Mhealth Uhealth

View full text Add to dashboard Cite

show abstract

“…Although the KoALA program, which is suitable for the Korean language, was used, 79 conversations included emoticons or URL video links and, thus, were excluded from text mining. As emoticons enrich the information provided on SNSs and strengthen social connections, further research to analyze emoticons is required, which will enhance usability [50][51][52]. Lastly, participation patterns in chats differed by group, and although positive effects through the SNS may occur, the potential for negative consequences (eg, loneliness and social comparison) cannot be excluded [47].…”

Section: Limitationsmentioning

confidence: 99%

Network Support Using Social Networking Services to Increase Exercise Adherence Among Korean-Chinese Middle-Aged Migrant Women: Mixed Methods Study (Preprint)

Lee¹,

Lee²,

Kim³

et al. 2020

Preprint

View full text Add to dashboard Cite

BACKGROUND Social networking services (SNSs) are recognized to be a promising approach to easily deliver health interventions and to enhance social support for exercise adherence. However, the patterns and aspects of social support through SNSs have not been reported and their influence on other social cognitive factors remain inconclusive. OBJECTIVE To explore how social support delivered through SNSs impacted interactions among KC middle-aged women and identify how this approach influenced social-cognitive factors for exercise (e.g., sense of community, self-efficacy for exercise, social support for exercise). METHODS A mixed-methods design was used. SNS Messages(n=259) were collected from 4 chat groups of 24 KC women who participated in a 12-week walking intervention program between August and October 2018. Text analysis of the messages and text mining using KoALA software were conducted. Social-cognitive factors (e.g., sense of community, self-efficacy for exercise, social support for exercise) were assessed at baseline and after 12 weeks using a structured questionnaire. A comparison of social-cognitive factors at baseline and 12 weeks was conducted to identify any potential significant changes using the Wilcoxon signed rank test. RESULTS The individual average frequency of chatting via SNSs was 10.79 (Range: 0-34) and the most frequent type of social support through SNSs was network support (66.4%). The most common words extracted from the SNS were “Health,” “Exercise,” “Participation,” and “We.” Overall, the perceived levels of sense of community (P<.001) and social support for exercise (P=.002) were significantly increased at 12 weeks compared with baseline. Groups 1(P=.03) and 4(P=.03), who demonstrated the highest frequency of network support only experienced a significant increase in the level of sense of community. CONCLUSIONS By integrating these data and conducting a mixed method analysis, we observed that among the types of social support, network support was a key point for the promotion of social-cognitive factors increasing exercise adherence. Therefore, network support through group-based SNSs should be considered as a useful strategy to help vulnerable migrant populations make changes to exercise behaviors.

show abstract

A Method of Analyzing Sentiment Polarity of Multilingual Social Media: A Case of Korean-Chinese Languages

Cited by 2 publications

References 17 publications

Network Support Using Social Networking Services to Increase Exercise Adherence Among Korean-Chinese Middle-Aged Migrant Women: Mixed Methods Study

Network Support Using Social Networking Services to Increase Exercise Adherence Among Korean-Chinese Middle-Aged Migrant Women: Mixed Methods Study

Network Support Using Social Networking Services to Increase Exercise Adherence Among Korean-Chinese Middle-Aged Migrant Women: Mixed Methods Study (Preprint)

Contact Info

Product

Resources

About