Analysis of Public Sentiment on Child Development Impact of Divorce With Naïve Bayes Algorithm Data Classifier on Twitter Using Rapidminer

Abstract

Divorce is a social phenomenon that can affect children's psychological, emotional and social development.  Social media such as Twitter has developed into an important platform for people to share their opinions and experiences on these issues in today's internet era.  As a result, an analysis technique is needed that can automatically process and classify people's feelings. This study uses the Naive Bayes Classifier algorithm to analyze public sentiment regarding child development caused by divorce.  The data used is taken from Twitter via scraping, which then goes through preprocessing processes such as cleaning, tokenization and stemming. This makes the data cleaner and ready for analysis.  Next, the data is categorized based on sentiment: positive, negative, and neutral. RapidMiner software is used to carry out the analysis process, which makes it easier to use the Naïve Bayes algorithm in sentiment classification. The research results show that the Naïve Bayes algorithm has a fairly high level of accuracy in classifying public sentiment regarding the impact of divorce on children.  The results of this analysis show that public opinion tends to be more negative, indicating concern about the condition of children after divorce. This research hopes to help governments, psychologists and social institutions create better policies and programs to help children affected by divorce.