Download PDFOpen PDF in browser

Twitter Sentiment Analysis on Long Acting Contraceptive Methods in Indonesia With Machine Learning Approach

EasyChair Preprint no. 10877

18 pagesDate: September 10, 2023


High population growth rates still need to be solved in almost all parts of the world. Over the past decade, Indonesia has been ranked fourth in the world regarding population. The family planning program has produced positive results. However, changes in the organizational structure of the institution and changes in local government commitment have led to high disparities in the family planning program. Long-term contraception is considered adequate as the primary need to reduce the rate of population increase. However, the pattern of contraceptive selection in couples of childbearing age in Indonesia is still dominated by non-long-term contraceptive methods such as pills and injections. In contrast, the rate of long-term contraceptive use continues to decline yearly. This study aims to see how the Indonesian people respond to long-term contraceptive products from January 1, 2020, to November 11, 2022. The research used data scraping techniques through the Twitter social media application by grouping sentiment based on negative, positive, and neutral classifications. Each sentiment was analyzed with a word cloud using keywords related to long-term contraceptive methods. Furthermore, classification evaluation is carried on by examining the accuracy of machine learning classification algorithms, specifically naive bayes and random forest.The results stated that the Indonesian people's response to long-term contraceptive methods is still negative, which means that long-term contraceptive products still need to be in demand by the Indonesian people. Based on the accuracy results, the random forest algorithm is very good at classifying tweets, which is 99,33% compared to the naïve bayes algorithm.

Keyphrases: Contraception, LARC, machine learning, Naive Bayes, Random Forest, Tweet, Twitter

BibTeX entry
BibTeX does not have the right entry for preprints. This is a hack for producing the correct reference:
  author = {Nurul Sari and Ahmad Munir and Muhammad Ramli and Madris Iskandar},
  title = {Twitter Sentiment Analysis on Long Acting Contraceptive Methods in Indonesia With Machine Learning Approach},
  howpublished = {EasyChair Preprint no. 10877},

  year = {EasyChair, 2023}}
Download PDFOpen PDF in browser