Data analytics of social media publicity to enhance household waste management

被引:57
作者
Jiang, Peng [1 ]
Fan, Yee Van [2 ]
Klemes, Jiri Jaromir [2 ]
机构
[1] ASTAR, Dept Syst Sci, Inst High Performance Comp, Singapore 138632, Singapore
[2] Brno Univ Technol VUT Brno, Fac Mech Engn, NETME Ctr, Sustainable Proc Integrat Lab SPIL, Tech 2896-2, Brno 61669, Czech Republic
基金
中国国家自然科学基金;
关键词
Digital waste management; User engagement; Internet of Things; Publicity improvement; Text data mining; SOLID-WASTE; FOOD WASTE; REGRESSION;
D O I
10.1016/j.resconrec.2020.105146
中图分类号
X [环境科学、安全科学];
学科分类号
08 ; 0830 ;
摘要
Household waste segregation and recycling is ranked at a high priority of the waste management hierarchy. Its management remains a great challenge due to the high dependency on social behaviours. The integration of Internet of Things (IoT) and subscription accounts on social media platforms related to household waste management could be an effective and environmentally friendly publicity approach than traditional publicity via posters and newspapers. However, there is a paucity of literature on measuring social media publicity in household waste management, which brings challenges for practitioners to characterise and improve this publicity pathway. In this study, under an integrated framework, data mining approaches are employed or extended for multidimensional publicity analytics using the data of online footprints of propagandist and users. A real-world case study based on a subscription account on the WeChat platform, Shanghai Green Account, is analysed to reveal useful insights for personalised improvements of household waste management. This study suggests that the current publicity related to household waste management leans towards propagandist-centred in both timing and topic dimensions. The identified timing, which has high user engagement, is 12:00-13:00 and 21:00-22:00 on Thursday. The overall relative publicity quality of historical posts is calculated as 0.95. Average user engagement under the macro policy in Shanghai was elevated by 138.5% from 2018 to 2019, during which the collections of biodegradable food waste and recyclable waste were elevated by 88.8% and 431.8%. Intelligent decision support by publicity analytics could enhance household waste management through effective communication.
引用
收藏
页数:12
相关论文
共 61 条
[1]   What is wrong with topic modeling? And how to fix it using search-based software engineering [J].
Agrawal, Amritanshu ;
Fu, Wei ;
Menzies, Tim .
INFORMATION AND SOFTWARE TECHNOLOGY, 2018, 98 :74-88
[2]  
Akaike H., 2011, International Encyclopedia of Statistical Science, p25), DOI [DOI 10.1007/978-3-642-04898-2110, DOI 10.1007/978-3-642-04898-2_110, 10.1007/978-3-642-04898-2110]
[3]  
Aldous Kholoud Khalil, 2019, Proceedings of the International AAAI Conference on Web and Social Media, V13, P47, DOI 10.1609/icwsm.v13i01.3208
[4]  
[Anonymous], 2017, P 9 INT C MACH LEARN
[5]   Latent Dirichlet allocation [J].
Blei, DM ;
Ng, AY ;
Jordan, MI .
JOURNAL OF MACHINE LEARNING RESEARCH, 2003, 3 (4-5) :993-1022
[6]  
Cole K., 2016, WASTE ADV MAG
[7]  
Cui R., 2018, PROD OPER MANAG, V27, P1749, DOI DOI 10.1111/poms.12707
[8]  
DEEP, 2020, US SOC MED PROM REC
[9]   Bayesian robustness to outliers in linear regression and ratio estimation [J].
Desgagne, Alain ;
Gagnon, Philippe .
BRAZILIAN JOURNAL OF PROBABILITY AND STATISTICS, 2019, 33 (02) :205-221
[10]   Measuring Skewness: A Forgotten Statistic? [J].
Doane, David P. ;
Seward, Lori E. .
JOURNAL OF STATISTICS EDUCATION, 2011, 19 (02)