Predicting election popularity of a person using crowd sensing in social networks
Abstract
Predicting election popularity from social network data is an appealing research topic. This covers all aspects from data collection to data representation through data processing. Although social media may provide a glimpse on electoral outcomes current research does not provide strong evidence to support it can replace traditional polls. Data scrapping could help us with crawling the data and create a database regarding that statistics which can predict the winner. We propose that social networking sites can provide an “open” publish-subscribe infrastructure to sense crowd and efficiently predict an election result for a political party or a political leader. The possibility of winning for a candidate will be predicted by mining representative terms from the social media that people posted before the election or during campaign. Such systems like crowd sensing can cause benefit to both the voters and the nominees. We are working on Twittter as our social media.