Subject: air pollution, pm10, sentiment analysis, cross correlation, topic modeling, fisher’s z-test, news media, tweets analysis
Year: 2023
Type: Proceeding article
Title: Comparative analysis of Air Pollution-related Tweets and News Article Teasers
Author: Perikj, Teodora
Author: Dedinec, Aleksandra
Author: Prodanova, Jana
Abstract: Air pollution is a consequence of both natural sources and human activities in the environment, which have a negative impact on the atmosphere. Air pollution is a serious environmental problem and poses a serious threat to the health of people in Macedonia. The purpose of the study is to investigate the society’s awareness of air pollution in Macedonia, compared with the findings from previous research. This study’s objective is to explore whether there is a correlation between tweets related to air pollution, teasers related to air pollution, and measured values of PM10 particles and if there is a difference in society’s awareness of air pollution compared to last year’s situation. To analyze our assumptions, we used the Natural Language Processing techniques: sentiment analysis and topic modeling, along with statistical analysis, fisher’s z-test and correlation analysis. The obtained results show us what feelings people express towards air pollution, what topics they talk about and help us determine if society's awareness has increased since last year.
Publisher: Ss Cyril and Methodius University in Skopje, Faculty of Computer Science and Engineering, Republic of North Macedonia
Relation: CIIT 2023 papers;17;
Identifier: oai:repository.ukim.mk:20.500.12188/27393
Identifier: http://hdl.handle.net/20.500.12188/27393
