User-Generated Content, Social Media, Less-resourced lan-guages, Sentiment translation
This paper presents two main methods of Sentiment Analysis(SA) of User-Generated Content for a low-resource language: Irish. Thefirst method, automatic sentiment translation, applies existing EnglishSA resources to both manually- and automatically-translated tweets. Weobtained an accuracy of 70% using this approach. The second method in-volved the manual creation of an Irish-language sentiment lexicon:Senti-Foclóir. This lexicon was used to build the first Irish SA system,Sen-tiFocalTweet, which produced superior results to the first method, withan accuracy of 76%. This demonstrates that translation from Irish toEnglish has a minor effect on the preservation of sentiment; it is alsoshown that theSentiFocalTweetsystem is a successful baseline systemfor Irish sentiment analysis
(4) Sentiment Translation for low resourced languages: Experiments on Irish General Election Tweets. Available from: https://www.researchgate.net/publication/315471653_Sentiment_Translation_for_low_resourced_languages_Experiments_on_Irish_General_Election_Tweets [accessed Apr 04 2018].