Monitoring the public opinion about the vaccination topic from tweets analysis

Abstract

The paper presents an intelligent system to automatically infer trends in the public opinion regarding the stance towards the vaccination topic: it enables the detection of significant opinion shifts, which can be possibly explained with the occurrence of specific social context-related events. The Italian setting has been taken as the reference use case. The source of information exploited by the system is represented by the collection of vaccine-related tweets, fetched from Twitter according to specific criteria; subsequently, tweets undergo a textual elaboration and a final classification to detect the expressed stance towards vaccination (i.e. in favor, not in favor, and neutral). In tuning the system, we tested multiple combinations of different text representations and classification approaches: the best accuracy was achieved by the scheme that adopts the bag-of-words, with stemmed n-grams as tokens, for text representation and the support vector machine model for the classification. By presenting the results of a monitoring campaign lasting 10 months, we show that the system may be used to track and monitor the public opinion about vaccination decision making, in a low-cost, real-time, and quick fashion. Finally, we also verified that the proposed scheme for continuous tweet classification does not seem to suffer particularly from concept drift, considering the time span of the monitoring campaign.

Vaccine Hesitancy on Social Media: Sentiment Analysis from June 2011 to April 2019

“…The vaccine hesitancy monitorisation could include text mining to extract data from social media messages, analyse it, classify the stance towards vaccination [ 9 ] and detect the main subtopics of concern. Sentiment analysis (SA) [ 10 ] is the text-mining subfield that allows the classification of opinions according to the polarity (positive, negative, or neutral) [ 11 ], the emotion (happiness, sadness, fear, etc.)…”

“…Basic information about Twitter and the definitions of the Twitter terms mentioned in this article can be found in Appendix A . Published articles about opinion analysis towards vaccination on social media usually perform short-term SA [ 9 , 24 , 25 ] in small datasets [ 24 , 26 ]. Some studies are limited to a single location [ 24 , 25 , 27 ], but usually geolocation is not analysed [ 3 ].…”

Incremental Learning of Fuzzy Decision Trees for Streaming Data Classification

“…Data stream analysis is gaining more and more attention thanks to the daily usage of a huge number of heterogeneous sources of streaming data such as sensors, wearables, smartphones and all other smart devices of the Internet of Things [17] in different application domains, including measurements in network monitoring and traffic management [6], log records and clickstreams in web surfing [2] as well as in virtual learning environments [12], manufacturing processes [19], and continuous monitoring of twitter posts [4]. Moreover, data stream analysis is part of the Big Data and Analytics enabling technology for Industry 4.0 [13].…”

Multi-class approach for user behavior prediction using deep learning framework on twitter election dataset

“…It provides TwitterAPI to crawl on tweets according to a particular hashtag (Cha et al, 2010;Trusov et al, 2010;Messias et al, 2013. Several approaches are already proposed for sentiment analysis using Twitter data such as unigram (Andrea et al, 2019), Word2Vec (Kapočiute-Dzikiene et al, 2019), Part-of-Speech (Agarwal et al, 2011), emoticon mining (Kumar & Sebastian, 2012), N-gram (Pak & Paroubek, 2010), parse tree (Martínez-Cámara et al, 2012;Saif et al, 2012).…”