FSAL: A Tailor-made Financial Sentiment Lexicon in Spanish for the Argentinian Markets (BYMA)

Abstract

During the last decade studies have shown that lexicon-based Sentiment Analysis of tweets combined with Machine Learning techniques can be used to enhance Algorithmic Trading strategies. The aim of the present work is to show how a specific domain lexicon in finance for the Argentinian Markets (FSAL) provides a better outcome than a generic lexicon (SDAL). First, we introduce a finance tailor-made lexicon. Secondly, we experimentally show that our lexicon outperforms a general purpose one on a corpus of tweets previously classified collaboratively by specialists in finance. Then, we compare the lexicons applying three different Machine Learning algorithms. Finally, we introduce some preliminary results and conclusions