Text Analysis of the Hybrid Digital Corpora

Digital means of communication have become the primary source of information for the majority of population. Given their availability through various platforms digital contents have an important role in shaping of the public interests and opinion. Every now and then a part of it becomes occupied with a new concept that resonates for a certain time, creating corpora of contents that persist on the platforms readily available for the consumers to reach them. It is critical to objectively determine what kind of discourse these text corpora created. In order to examine one such case, related to the use of the term "hibridni rat" (en. hybrid war), this research conducted a study that applied contemporary text analysis techniques. The process consisted of several stages: initially, having investigated the research phenomenon, the analytical problems were defined. This was followed by the identification of the digital contents posted under the national domain of the Republic of Croatia, their retrieval, structuring and pre-processing. Thus, the digital corpora suitable for analysis were created and they were subjected to processing by using the text mining techniques. The results of this process have provided an enhanced insight into the relatively large corpora of digital texts that are difficult for a reader to review, grasp on, and extract the useful information about the phenomenon being investigated just by using the traditional means of browsing the contents. The proposed approach for text fetching and analysis is a major contribution of this study. At the same time the described procedure is relatively easy to reproduce so it can be used in the analysis of texts available in digital format that are related to other phenomena.