An Application of Machine Learning on Social Media for Solid Waste Management

• Published in: 2024 5th International Conference on Artificial Intelligence and Data Sciences, AiDAS 2024 - Proceedings
• Main Author: Januri S.S.; Rozi N.A.B.M.; Nasir N.; Mustaf-Jab S.N.; Sahar N.N.
• Format: Conference paper
• Language: English
• Published: Institute of Electrical and Electronics Engineers Inc. 2024
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85209651898&doi=10.1109%2fAiDAS63860.2024.10730401&partnerID=40&md5=e0cd1ff000fee5fb97e5c3bebf357bc2

The purpose of this study is to analyze the social media publicity of solid waste management through a Twitter page. The objectives are to determine the sentiment and frequently used keywords related to solid waste management on Twitter, and to analyze the ranking order and contribution levels of these topics using sentiment analysis, the Bag-of-N-Grams model, and Latent Dirichlet Allocation (LDA). The results show that the Naïve Bayes model achieves the highest accuracy (84.79%). The Bag-of-N-Grams model reveals that tweets frequently use words such as kitar, sisa, and bersih to raise public awareness about solid waste management, with frequencies of 260, 214, and 213, respectively. An analysis of the ranking order of topics shows that Topic 7 (Pasca Banjir Selangor) has the highest probability of being posted, with a probability of 0.7030. Nevertheless, the frequency of posting all topics regarding the solid waste management appear to be greater than 50%. It can be seen the utilization of the machine learning technique yields valuable insights to improve waste management practices. © 2024 IEEE.