Monitoring Environmental Quality by Sniffing Social Media
Open Access
- 10 February 2017
- journal article
- research article
- Published by MDPI AG in Sustainability
- Vol. 9 (2) , 85
Abstract
Nowadays, the environmental pollution and degradation in China has become a serious problem with the rapid development of Chinese heavy industry and increased energy generation. With sustainable development being the key to solving these problems, it is necessary to develop proper techniques for monitoring environmental quality. Compared to traditional environment monitoring methods utilizing expensive and complex instruments, we recognized that social media analysis is an efficient and feasible alternative to achieve this goal with the phenomenon that a growing number of people post their comments and feelings about their living environment on social media, such as blogs and personal websites. In this paper, we self-defined a term called the Environmental Quality Index (EQI) to measure and represent people’s overall attitude and sentiment towards an area’s environmental quality at a specific time; it includes not only metrics for water and food quality but also people’s feelings about air pollution. In the experiment, a high sentiment analysis and classification precision of 85.67% was obtained utilizing the support vector machine algorithm, and we calculated and analyzed the EQI for 27 provinces in China using the text data related to the environment from the Chinese Sina micro-blog and Baidu Tieba collected from January 2015 to June 2016. By comparing our results to with the data from the Chinese Academy of Sciences (CAS), we showed that the environment evaluation model we constructed and the method we proposed are feasible and effective.Keywords
This publication has 15 references indexed in Scilit:
- Using geosocial search for urban air pollution monitoringPervasive and Mobile Computing, 2017
- A hybrid model of sentimental entity recognition on mobile social mediaEURASIP Journal on Wireless Communications and Networking, 2016
- Sensing Pollution on Online Social Networks: A Transportation PerspectiveMobile Networks and Applications, 2016
- Chinese social media analysis for disease surveillancePersonal and Ubiquitous Computing, 2015
- Can Social Media Clear the Air? A Case Study of the Air Pollution Problem in Chinese CitiesThe Professional Geographer, 2014
- Inferring air pollution by sniffing social media2014
- Socioscope: Spatio-temporal Signal Recovery from Social Media2012
- Evaluation of simple performance measures for tuning SVM hyperparametersNeurocomputing, 2003
- A Tutorial on Support Vector Machines for Pattern RecognitionData Mining and Knowledge Discovery, 1998
- Support-vector networksMachine Learning, 1995