Abstract
The study introduces latest research findings from the area of machine processing of Czech emotional language data. First, it provides an analysis of language means which together form an emotional meaning of written utterances in Czech.
Second, it employs the findings concerning emotional language in computational applications. The author provides a systematic overview of lexical, morphosyntactic, semantic and pragmatic aspects of emotional meaning in Czech utterances.
Also, she proposes two formal representations of emotional structures within the framework of the Prague Dependency Treebank and Construction Grammar. Regarding the computational applications, the study focuses on sentiment analysis, i.e. automatic extraction of emotions from text.
It describes a creation of manually annotated emotional data resources in Czech and performs main sentiment analysis tasks, such as e.g. polarity classification and opinion target identification on Czech data, employing the up-to-date methods of natural lang