Implementing Sentiment Dictionary into Vader Sentiment Analyis for Croatian

Autor: Ban Kirigin, Tajana, Bujačić Babić, Sanda, Perak, Benedikt
Jazyk: angličtina
Rok vydání: 2021
Předmět:
Popis: We describe the processes of 1) sequential propagation of sentiment values of nouns, adjectives, verbs and proper nouns from the multilingual SenticNet 6 sentiment dictionary using ENGRI corpus coordination syntactic- dependency ; 2) extrapolation of lemma values to word forms ; 3) implementation of the generated sentiment dictionary and Vader rule- based logic on the grammatical structure of Croatian sentence sentiment analysis. Although Vader has been modified to work with multiple languages using translation and adapted in several languages, to the best of our knowledge, this is the first attempt to implement a dictionary resource and adapt the grammar logic to a rule- based sentiment analysis tool Vader for the Croatian language. The work presented in this paper aims at providing the research community with a tool that can be used to extend the range of sentiment analysis tools and resources for the Croatian language.
Databáze: OpenAIRE