Abstract | ||
---|---|---|
Several studies have shown how to approximately predict public opinion,such as in political elections, by analyzing user activities in blogging platformsand on-line social networks. The task is challenging for several reasons.Sample bias and automatic understanding of textual content are two of severalnon trivial issues.In this work we study how Twitter can provide some interesting insights concerningthe primary elections of an Italian political party. State-of-the-art approachesrely on indicators based on tweet and user volumes, often including sentimentanalysis. We investigate how to exploit and improve those indicators in order toreduce the bias of the Twitter users sample. We propose novel indicators and anovel content-based method. Furthermore, we study how a machine learning approachcan learn correction factors for those indicators. Experimental results onTwitter data support the validity of the proposed methods and their improvementover the state of the art. |
Year | Venue | Field |
---|---|---|
2015 | IIR | Data science,Data mining,Political Elections,Social network,Primary election,Computer science,Sampling bias,Artificial intelligence,World Wide Web,Sentiment analysis,Exploit,Public opinion,Politics,Machine learning |
DocType | Citations | PageRank |
Conference | 4 | 0.45 |
References | Authors | |
6 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Mauro Coletto | 1 | 65 | 5.74 |
Claudio Lucchese | 2 | 1104 | 73.76 |
Salvatore Orlando | 3 | 1595 | 202.29 |
Raffaele Perego | 4 | 1471 | 108.91 |