|
|
Virtual Image of the Czech Sports via Unstructured Data Analysis
Levý, Jan ; Jelínek, Ivan (advisor) ; Profousová, Lenka (referee)
Analysis of unstructured data from social media is an important and constantly growing part of information technologies' application in the marketing context. The aims of this bachelors' thesis include creating a trial platform for the analysis of un-structured data from Facebook, research and application of suitable methods and metrics for the analysis, summarizing the analysis' findings and visualisation of selected metrics by means of a dashboard. The thesis' structure corresponds to the methods used for metting the aims set and compo-ses of seven parts. The first part summarizes motives for social media analysis, deals with the approach to the topic in the contemporary scientific papers and discusses comercial tools that could be used for such analysis. The second part describes gathering of data with a crawler, setting the connection and getting the access rights to the data sources. The structure of downloaded documents and appropriate choice of data sources is also described in this part. The third part adresses the tools used for the analysis, namely Elas-ticsearch and Kibana. The fourth part defines possible questions that may be asked in the context of Czech sports image on Facebook. The fifth part identifies the metrics necessary for the questions assesment and describes corresponding data selection. The sixth part completes the analysis itself and summarizes its findnings. The seventh part describes Ki-bana's dashboard use for effective summarizing of selected metrics. The main bachelors' thesis contribution lies in the illustration of social media analysis pos-sibilities on the example of Czech sportsmen and Czech sports in general on Facebook. A crawler has been used to gather the data for the purpose of the analysis. The other em-ployed tools included Elasticsearch and Kibana, which have enabled the data selection and visualisation. The analytical approach consisted of four parts; definition of analytical que-stions, definition of methods and metrics to find answers to these questions, followed by the analysis itself and the summary of results. The following points belong to the analysis' findnings: identification of sportsman with the highest marketing potential, comparison of analysed sports according to the perceived image by fans and desgination of the pages with most interacting followers.
|
guest ::
