submitted to the 5th International Conference on Data Analytics, IARIA 2016, Venice, Italy, Oct 9-13, 2016
The article explores the impact of Machine Translation on sentiment analysis, employing the combination of two state-ofthe- art tools – the multilingual sentiment analysis tool SentiSAIL and the Machine Translation tool SDL Language Weaver. The original corpora are in German, Russian and Spanish in the domain of general news. The output language of translation is English. Firstly, the work presents the development and evaluation of SentiSAIL features in a newly supported language – Spanish. Further experimental setup reveals that the performance rates of sentiment analysis on the original and translated corpora are comparable. Thus a given tool, performing high quality Machine Translation from a target language to English, can eliminate the necessity to develop specific sentiment analysis resources for that language.
To access the full article, please fill in the form below.
Request Form for Research Paper