{"title":"SENTIMENT ANALYSIS OF COMMENTS ON SEXUAL HARASSMENT IN COLLEGES ON FOUR POPULAR SOCIAL MEDIA","authors":"Vinson Phoan, Johan Setiawan","doi":"10.53748/jmis.v2i2.33","DOIUrl":null,"url":null,"abstract":"Sexual harassment is a sexual act in the form of verbal and nonverbal that is carried out intentionally, and there are indications of coercion on the victim. Often the victim refuses to report or tell stories because the victim is afraid that something untoward will happen in the future. To find helpful information for survivors and research in this study, we will get the sentiment on comments related to matters related to UMN by using the CRISPDM framework method, FastText, and the SVM algorithm to identify a statement on sexual relations with UMN.Results From this study used a model with an accuracy of 55.14% and sexual harassment data collected on March 16, 2022, with 287 data obtained from Twitter, Instagram, Medium, and Line Today sites. Positive sentiment is the least found sentiment from the overall data, only 8.7%, while negative sentiment is 36.6%. Of the four platforms, the best is the Twitter platform because it gets a pretty good response in terms of positive and neutral sentiments compared to others.","PeriodicalId":331767,"journal":{"name":"Journal of Multidisciplinary Issues","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2022-08-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Journal of Multidisciplinary Issues","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/10.53748/jmis.v2i2.33","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
引用次数: 0
Abstract
Sexual harassment is a sexual act in the form of verbal and nonverbal that is carried out intentionally, and there are indications of coercion on the victim. Often the victim refuses to report or tell stories because the victim is afraid that something untoward will happen in the future. To find helpful information for survivors and research in this study, we will get the sentiment on comments related to matters related to UMN by using the CRISPDM framework method, FastText, and the SVM algorithm to identify a statement on sexual relations with UMN.Results From this study used a model with an accuracy of 55.14% and sexual harassment data collected on March 16, 2022, with 287 data obtained from Twitter, Instagram, Medium, and Line Today sites. Positive sentiment is the least found sentiment from the overall data, only 8.7%, while negative sentiment is 36.6%. Of the four platforms, the best is the Twitter platform because it gets a pretty good response in terms of positive and neutral sentiments compared to others.