Sentiment Aggregate Functions for Political Opinion Polling using Microblog Streams

TL;DR

This paper evaluates various sentiment aggregation functions on microblog streams to improve political opinion polling accuracy, analyzing a large Portuguese tweet dataset to identify how different functions influence predictive features over time.

Contribution

It introduces a comprehensive analysis of sentiment aggregate functions and their impact on political opinion prediction from microblog data, highlighting temporal feature importance variations.

Findings

Different sentiment aggregate functions vary in feature importance over time.

The overall prediction error remains stable despite changes in feature importance.

Sentiment aggregation methods can effectively inform political opinion models.

Abstract

The automatic content analysis of mass media in the social sciences has become necessary and possible with the raise of social media and computational power. One particularly promising avenue of research concerns the use of sentiment analysis in microblog streams. However, one of the main challenges consists in aggregating sentiment polarity in a timely fashion that can be fed to the prediction method. We investigated a large set of sentiment aggregate functions and performed a regression analysis using political opinion polls as gold standard. Our dataset contains nearly 233 000 tweets, classified according to their polarity (positive, negative or neutral), regarding the five main Portuguese political leaders during the Portuguese bailout (2011-2014). Results show that different sentiment aggregate functions exhibit different feature importance over time while the error keeps almost unchanged.