Data and text mining in information management: A digital content analysis application

In this study, the raw data collected from an organization that publishes digital content was processed within the context of information management in the organization. Two separate analyses were carried out for the organization and valuable information was obtained as a result. In the first analysis, with the aim of finding out the factors affecting the reading rates for the published content, texts were analyzed with a particular text mining method and the factors determining the reading rates were revealed with the multiple linear regression analysis. Moreover, it was demonstrated that the particular text mining approach applied to the problem in this analysis is statistically valid. Additionally, a title based digital content analysis process was developed for this research. In the second analysis, the effect of the reading rates on the organization's income was evaluated. For this purpose, monthly company revenue and the reading counts were considered as a simple linear regression problem and an estimation model was established. Finally, an attempt was made to explain all the processes, from the acquisition of the raw data to the creation of variables with data processing, as well as the analyses and the interpretation of the results. Thus, the thesis aims to be an example for the application of end-to-end data mining in the field of information management.