NRC-Canada: Building the State-of-the-Art in Sentiment Analysis of\n Tweets

This paper describes how it created two state-of-the-art SVM classifiers, one to detect the sentiment of messages such as tweets and SMS (message-level task) and one to detects the sentimentof a term within a message (term-leveltask).

Abstract

In this paper, we describe how we created two state-of-the-art SVM\nclassifiers, one to detect the sentiment of messages such as tweets and SMS\n(message-level task) and one to detect the sentiment of a term within a\nsubmissions stood first in both tasks on tweets, obtaining an F-score of 69.02\nin the message-level task and 88.93 in the term-level task. We implemented a\nvariety of surface-form, semantic, and sentiment features. with sentiment-word\nhashtags, and one from tweets with emoticons. In the message-level task, the\nlexicon-based features provided a gain of 5 F-score points over all others.\nBoth of our systems can be replicated us available resources.\n