Publication detail

Data analysis of resident engagement and sentiments in social media enables better household waste segregation and recycling

Jiang, P. Zhou, J. Fan, Y.V. Klemeš, J.J. Zheng, M. Varbanov, P.S.

English title

Data analysis of resident engagement and sentiments in social media enables better household waste segregation and recycling

Type

journal article in Web of Science

Language

en

Original abstract

Waste segregation, recycling and reduction have been prioritised in the Circular Economy transition of household waste management to reduce environmental impacts. With digitalisation and innovation developments in waste management, residents become more active on waste management-related social media platforms. However, there is still needed a tangible analysis of resident engagement (e.g. user comments and interactions) and related sentiment changes on such platforms to enhance waste management and ease the environmental burden at source. This study develops an integrated solution to analyse resident engagement by leveraging statistical analysis and text-mining methods. Four interrelated components are incorporated in the solution: population behaviour quantification, sentiment analysis and dynamics, popular concerns and probability distribution fitting, and rule-based managerial insight identification. The novel solution is applied to a real-world case study on a subscription account related to waste management in Shanghai. This research produces several major observations based on the studied case: (i) The resident engagement Monday-to-Thursday was more active than Friday-to-Sunday. (ii) Compared to 2018, the resident engagement by commenting on online posts was elevated by 107.1% in 2019 when Shanghai introduced a new management policy. Meanwhile, the yearly resource-type waste collection was increased by 114.5% in 2019. (iii) It took approximately one year to recover positive sentiments in user comments after introducing the policy. However, the comments with negative sentiments assisted in improving waste management. (iv) The best-fitted negative binomial distribution of the number of votes for user comments could guarantee the managerial insight identification from the minority of comments with popular concerns.

English abstract

Waste segregation, recycling and reduction have been prioritised in the Circular Economy transition of household waste management to reduce environmental impacts. With digitalisation and innovation developments in waste management, residents become more active on waste management-related social media platforms. However, there is still needed a tangible analysis of resident engagement (e.g. user comments and interactions) and related sentiment changes on such platforms to enhance waste management and ease the environmental burden at source. This study develops an integrated solution to analyse resident engagement by leveraging statistical analysis and text-mining methods. Four interrelated components are incorporated in the solution: population behaviour quantification, sentiment analysis and dynamics, popular concerns and probability distribution fitting, and rule-based managerial insight identification. The novel solution is applied to a real-world case study on a subscription account related to waste management in Shanghai. This research produces several major observations based on the studied case: (i) The resident engagement Monday-to-Thursday was more active than Friday-to-Sunday. (ii) Compared to 2018, the resident engagement by commenting on online posts was elevated by 107.1% in 2019 when Shanghai introduced a new management policy. Meanwhile, the yearly resource-type waste collection was increased by 114.5% in 2019. (iii) It took approximately one year to recover positive sentiments in user comments after introducing the policy. However, the comments with negative sentiments assisted in improving waste management. (iv) The best-fitted negative binomial distribution of the number of votes for user comments could guarantee the managerial insight identification from the minority of comments with popular concerns.

Keywords in English

Data analysis; Digitalised waste management; Increased waste recycling; Resident engagement; Social media; Waste reduction

Released

15.10.2021

Publisher

Elsevier, Ltd.

Location

ELSEVIER SCI LTD, THE BOULEVARD, LANGFORD LANE, KIDLINGTON, OXFORD OX5 1GB, OXON, ENGLAND

ISSN

0959-6526

Number

319

Pages from–to

128809–128809

Pages count

12

BIBTEX


@article{BUT172464,
  author="Yee Van {Fan} and Jiří {Klemeš} and Petar Sabev {Varbanov},
  title="Data analysis of resident engagement and sentiments in social media enables better household waste segregation and recycling",
  year="2021",
  number="319",
  month="October",
  pages="128809--128809",
  publisher="Elsevier, Ltd.",
  address="ELSEVIER SCI LTD, THE BOULEVARD, LANGFORD LANE, KIDLINGTON, OXFORD OX5 1GB, OXON, ENGLAND",
  issn="0959-6526"
}