24 March 2020
In the past few months, the literary studies subprojects in hermA have worked intensively on carrying out a sentiment analysis of fictional texts and adapting the methodology to literary studies related questions. Due to complex negations, ironic and metaphorical ways of speaking, unreliable results are often generated. Therefore we have developed a method that measures the 'sentiment sensitivity' in the characters' speeches ignoring the polarity of sentiment lemmas.
For the worldwide conference of Digital Humanities (DH 2020) the paper "Sentiment Sensitivity: Using sentiment analysis in literary studies to analyze genre and the depiction of illness" was accepted. hermA is thus represented at the international DH conference for the second time. Last year the paper "Evaluation of a Semantic Field-Based Approach to Identifying Text Sections about Specific Topics" was presented.