PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
April 11, 2026Waste Management Bulletin0 citationsOpen Access

Extraction of waste economic variables through cluster analysis from web scraping results and latent Dirichlet allocation

View Full Paper
SSolimunAFAdji Achmad Rinaldo FernandesFJFachira Haneinanda Junianto

Key Points

  • The aim is to map societal views on the waste economy and identify opportunities for behavioral change.
  • Web scraping from Twitter collected 3,964 reviews about waste management.
  • Latent Dirichlet Allocation (LDA) identified 18 thematic topics related to waste.
  • Cluster analysis extracted structured waste economy variables from unstructured text.
  • High public awareness of recycling and waste banks, with a score of 4.23.
  • Limited understanding of reduction and reuse principles was noted.
  • A strong economic perception of waste was supported by a score of 4.44.

Abstract

• A novel approach using web scraping and text mining (LDA) is applied to assess public perception of the waste economy. • 18 key thematic topics related to waste management and its economic value were identified from 3,964 social media reviews. • Cluster analysis was used to extract structured waste economy variables from unstructured textual data. • Public perception shows high awareness of recycling and waste bank usage but limited understanding of reduce and reuse principles. • Findings suggest that waste has strong economic potential, supporting the development of sustainable waste-based economic models. This study explores public perceptions of Batu City, Indonesia the waste economy and waste management as an economic resource. The research aims to map societal views on waste as an economic resource, highlighting opportunities for behavioral change and sustainable practices. The novelty of this study lies in integrating digital public discourse with waste economy concepts, offering a socio-digital perspective that extends beyond purely technical or policy-oriented approaches. The findings emphasize the need for improved public education on reduction and reuse and for strengthening waste management infrastructure, providing practical insights for policy development, community empowerment, and the advancement of sustainable green economy strategies. Data were collected through web scraping from Twitter platform throughout 2023, within a one-year research period, yielding 3,964 reviews. After preprocessing (case folding, tokenizing, spelling normalization, and stopword removal), 630 cleaned reviews were analyzed using Latent Dirichlet Allocation (LDA), which identified 18 key topics. These topics include environmental awareness, the role of waste banks, and the application of the 3R principles: Reduce, Reuse, and Recycle. The results indicate a relatively high level of public awareness, particularly in supporting waste banks and recycling efforts, which received the highest score of 4.23. However, the aspects of reduction and reuse still require further education. While cleaning facilities are considered sufficient, their maintenance remains suboptimal. The findings also show a growing public recognition of waste’s economic potential, supported by a 4.44 score on the indicator for utilizing waste as an economic resource. The study concludes that the waste economy presents promising opportunities for advancing environmental and economic sustainability. To maximize these benefits, efforts should be focused on improving public education regarding reduction and reuse, along with enhancing the effectiveness and maintenance of waste management infrastructure. This research is limited to text-based data from Twitter, analyzed using the LDA model, with findings influenced by the coherence score applied. Moreover, the scope is centered on issues such as environmental awareness, waste banks, and 3R implementation, without addressing the technical or policy aspects of waste management in greater depth.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Solimun et al. (2026) studied this question.

synapsesocial.com/papers/69d9e47378050d08c1b751b4https://doi.org/10.1016/j.wmb.2026.100302
Ask AI
Helpful
Bookmark
Share
View Full Paper