Modelling spatiotemporal variation of positive and negative sentiment on Twitter to improve the identification of localised deviations

02/22/2018
by   Zubair Shah, et al.
0

Studies examining how sentiment on social media varies over time and space appear to produce inconsistent results. Analysing 16.54 million English-language tweets from 100 cities posted between 13 July and 30 November 2017, our aim was to clarify how spatiotemporal and social factors contributed to variation in sentiment on Twitter. We estimated positive and negative sentiment for each of the cities using dictionary-based sentiment analysis and constructed models to explain differences in sentiment using time of day, day of week, weather, interaction type (social or non-social), and city as factors. Tests in a distinct but contiguous period of time showed that all factors were independently associated with sentiment. In the full multivariable model of positive (Pearson's R in test data 0.236; 95 (Pearson's R in test data 0.306 95 day explained more of the variance than other factors. Extreme differences between observed and expected sentiment using the full model appeared to be better aligned with international news events than degenerate models. In applications that aim to detect localised events using the sentiment of Twitter populations, it is useful to account for baseline differences before looking for unexpected changes.

READ FULL TEXT
research
09/14/2015

Twitter Sentiment Analysis

This project addresses the problem of sentiment analysis in twitter; tha...
research
09/05/2019

A Discussion on Influence of Newspaper Headlines on Social Media

Newspaper headlines contribute severely and have an influence on the soc...
research
02/01/2022

Politics and Virality in the Time of Twitter: A Large-Scale Cross-Party Sentiment Analysis in Greece, Spain and United Kingdom

Social media has become extremely influential when it comes to policy ma...
research
04/11/2016

In the mood: the dynamics of collective sentiments on Twitter

We study the relationship between the sentiment levels of Twitter users ...
research
06/30/2020

Twitter and Census Data Analytics to Explore Socioeconomic Factors for Post-COVID-19 Reopening Sentiment

Investigating and classifying sentiments of social media users towards a...
research
04/05/2017

Learning to Generate Reviews and Discovering Sentiment

We explore the properties of byte-level recurrent language models. When ...
research
11/19/2015

Overcoming Language Variation in Sentiment Analysis with Social Attention

Variation in language is ubiquitous, particularly in newer forms of writ...

Please sign up or login with your details

Forgot password? Click here to reset