In this thesis a subject mapping data mining method has been developed. The method maps the 8 political parties in the Swedish parliament to 25 subjects based on their participation in the subjects on Twitter. The method is based on tweets written by the commissioners in the Swedish parliament and the ministers in the Swedish government between 2018-04-01 and 2019-04-01. The method shows what subjects each party participate in on Twitter and how the participation changes over time. Further, it shows the similarity and dynamics between the parties according to their subject participation. The purpose of the method is to offer a quantitative analysis tool, useful to social scientist, that provides new information about political parties.It can not be concluded how the resulting findings from the method should be interpreted. However, I present three questions to investigate in future work. The questions regard correlation between the subject map and party cooperation, political spectra and public statements.The result indicate that there is a significance issue, due to low participation in certain subjects by specific parties. This issue can be resolved by decreasing the number of subjects or expand the data set on which the subject mapping is based.
| Identifer | oai:union.ndltd.org:UPSALLA1/oai:DiVA.org:uu-388311 |
| Date | January 2019 |
| Creators | Norberg, Anton |
| Publisher | Uppsala universitet, Avdelningen för datalogi |
| Source Sets | DiVA Archive at Upsalla University |
| Language | English |
| Detected Language | English |
| Type | Student thesis, info:eu-repo/semantics/bachelorThesis, text |
| Format | application/pdf |
| Rights | info:eu-repo/semantics/openAccess |
| Relation | UPTEC F, 1401-5757 ; 19024 |
Page generated in 0.0017 seconds