Building a Twitter Sentiment Analysis System with Recurrent Neural Networks

Nistor, Sergiu Cosmin; Moca, Mircea; Moldovan, Darie; Oprean, Delia Beatrice; Nistor, Răzvan

doi:10.3390/s21072266

Cited by 26 publications

(10 citation statements)

References 31 publications

(40 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using the Twitter API for Academic Research [ 26 ], we will collect the timestamp of each returned tweet and analyze its text for sentiment (positive or negative) by integrating with the Natural Language Toolkit in Python and other techniques such as latent Dirichlet allocation [ 27 ], Sentence BERT [ 24 ], or recurrent neural networks [ 28 ]. We will also analyze the number and types of health effects mentioned in association with cannabis use in pregnancy and will extract location data when available from each tweet, either from geotagged tweets or from the location associated with the user’s profile [ 29 ].…”

Section: Methodsmentioning

confidence: 99%

The Volume and Tone of Twitter Posts About Cannabis Use During Pregnancy: Protocol for a Scoping Review

Cresswell¹,

Espín-Noboa²,

Murphy³

et al. 2022

JMIR Res Protoc

View full text Add to dashboard Cite

Background Cannabis use has increased in Canada since its legalization in 2018, including among pregnant women who may be motivated to use cannabis to reduce symptoms of nausea and vomiting. However, a growing body of research suggests that cannabis use during pregnancy may harm the developing fetus. As a result, patients increasingly seek medical advice from online sources, but these platforms may also spread anecdotal descriptions or misinformation. Given the possible disconnect between online messaging and evidence-based research about the effects of cannabis use during pregnancy, there is a potential for advice taken from social media to affect the health of mothers and their babies. Objective This study aims to quantify the volume and tone of English language posts related to cannabis use in pregnancy from January 2012 to December 2021. Methods Modeling published frameworks for scoping reviews, we will collect publicly available posts from Twitter that mention cannabis use during pregnancy and use the Twitter Application Programming Interface for Academic Research to extract data from tweets, including public metrics such as the number of likes, retweets, and quotes, as well as health effect mentions, sentiment, location, and users’ interests. These data will be used to quantify how cannabis use during pregnancy is discussed on Twitter and to build a qualitative profile of supportive and opposing posters. Results The CHEO Research Ethics Board reviewed our project and granted an exemption in May 2021. As of December 2021, we have gained approval to use the Twitter Application Programming Interface for Academic Research and have developed a preliminary search strategy that returns over 3 million unique tweets posted between 2012 and 2021. Conclusions Understanding how Twitter is being used to discuss cannabis use during pregnancy will help public health agencies and health care providers assess the messaging patients may be receiving and develop communication strategies to counter misinformation, especially in geographical regions where legalization is recent or imminent. Most importantly, we foresee that our findings will assist expecting families in making informed choices about where they choose to access advice about using cannabis during pregnancy. Trial Registration Open Science Framework 10.17605/OSF.IO/BW8DA; www.osf.io/6fb2e International Registered Report Identifier (IRRID) PRR1-10.2196/34421

show abstract

Section: Methodsmentioning

confidence: 99%

The Volume and Tone of Twitter Posts About Cannabis Use During Pregnancy: Protocol for a Scoping Review

Cresswell¹,

Espín-Noboa²,

Murphy³

et al. 2022

JMIR Res Protoc

View full text Add to dashboard Cite

show abstract

“…Using the Twitter API for Academic Research [21], we will collect the timestamp of each returned tweet and analyze its text for sentiment (positive or negative) by integrating with the Natural Language Toolkit (NLTK) in Python and other techniques such as latent dirichlet allocation [22], Sentence-BERT [20], or recurrent neural networks [23]. We will also analyze the number and types of health effects mentioned in association with cannabis use in pregnancy and will extract location data where available from each tweet, either from geo-tagged tweets or from the location associated with the user's profile [24].…”

Section: Methodsmentioning

confidence: 99%

The Volume and Tone of Twitter Posts About Cannabis Use During Pregnancy: A Scoping Review Protocol

Cresswell¹,

Espín-Noboa²,

Murphy³

et al. 2021

Preprint

View full text Add to dashboard Cite

Background: Cannabis use has increased in Canada since its legalization in 2018, includingamong pregnant women who may be motivated to use cannabis to reduce symptoms ofnausea and vomiting. However, a growing body of research suggests that cannabis useduring pregnancy may harm the developing fetus. As a result, patients increasingly seekmedical advice from online sources, but these platforms may also spread anecdotaldescriptions or misinformation. Given the possible disconnect between online messaging andevidence-based research about the effects of cannabis use during pregnancy, there is apotential for advice taken from social media to cause harm.Objectives: To quantify the volume and tone of English-language posts related to cannabisuse in pregnancy from January 2012 to July 2021.Methods: Modelling published frameworks for scoping reviews, we will collect publiclyavailable posts from Twitter that mention cannabis use during pregnancy and employ theTwitter Application Programming Interface (API) for Academic Research to extract data fromtweets, including public metrics such as the number of likes, retweets and quotes, as well ashealth effect mentions, sentiment, location and users interests. These data will be used toquantify how cannabis use during pregnancy is discussed on Twitter and to build a qualitativeprofile of supportive and opposing posters.Results: The CHEO Research Ethics Board reviewed our project and granted an exemptionin May 2021. As of September 2021, we have gained approval to use the Twitter API forAcademic Research and have developed a preliminary search strategy that returns over 2million unique tweets posted between 2012 and 2020.Conclusions: Understanding how Twitter is being used to discuss cannabis use duringpregnancy will help public health agencies and healthcare providers assess the messagingpatients may be receiving and develop communication strategies to counter misinformation,especially in geographical regions where legalization is recent or imminent. Most importantly,we foresee that our findings will assist expecting families in making informed choices aboutwhere they choose to access advice about using cannabis during pregnancy.

show abstract

“…The hybrid approach of analyzing sentiments consisting the statistical and knowledge-based methods to recognize the polarity [37]. Most novel researchers are using a hybrid approach to sentiment analysis due to the reason that this method can enhance the accuracy of the sentiment analyzing model [38,39,40].…”

Section: Hybrid Approachmentioning

confidence: 99%

“…The Machine Learning based classifications are using numerous methods for data classification like SVM, NB, ME [46] etc. and as the novel methods RNN, CNN, LSTM, GRU [39,47,48]can be mentioned. The analyzed results can be evaluated and test the accuracy.…”

Section: Classification and Evaluationmentioning

confidence: 99%

Literature review on Real-time Location-Based Sentiment Analysis on Twitter

Rathnayaka¹,

Jayasena²,

Ratnayake³

2021

Adv. Technol.

View full text Add to dashboard Cite

Sentiment analysis mainly supports sorting out the polarity and provides valuable information with the use of raw data in social media platforms. Many fields like health, business, and security require real-time data analysis for instant decision-making situations.Since Twitter is considered a popular social media platform to collect data easily, this paper is considering data analysis methods of Twitter data, real-time Twitter data analysis based on geo-location. Twitter data classification and analysis can be done with the use of diverse algorithms and deciding the most appropriate algorithm for data analysis, can be accomplished by implementing and testing these diverse algorithms.This paper is discussing the major description of sentiment analysis, data collection methods, data pre-processing, feature extraction, and sentiment analysis methods related to Twitter data. Real-time data analysis arises as a major method of analyzing the data available online and the real-time Twitter data analysis process is described throughout this paper. Several methods of classifying the polarized Twitter data are discussed within the paper while depicting a proposed method of Twitter data analyzing algorithm. Location-based Twitter data analysis is another crucial aspect of sentiment analyses, that enables data sorting according to geo-location, and this paper describes the way of analyzing Twitter data based on geo-location. Further, a comparison about several sentiment analysis algorithms used by previous researchers has been reported and finally, a conclusion has been provided.

show abstract

Building a Twitter Sentiment Analysis System with Recurrent Neural Networks

Cited by 26 publications

References 31 publications

The Volume and Tone of Twitter Posts About Cannabis Use During Pregnancy: Protocol for a Scoping Review

The Volume and Tone of Twitter Posts About Cannabis Use During Pregnancy: Protocol for a Scoping Review

The Volume and Tone of Twitter Posts About Cannabis Use During Pregnancy: A Scoping Review Protocol

Literature review on Real-time Location-Based Sentiment Analysis on Twitter

Contact Info

Product

Resources

About