This paper describes the development of two state-of-the-art SVM classifiers for sentiment analysis of tweets and SMS messages. The first classifier detects the overall sentiment of a message (message-level task), while the second classifier identifies the sentiment of individual terms within a message (term-level task). The authors participated in the SemEval-2013 competition and achieved the highest scores in both tasks, with an F-score of 69.02 for the message-level task and 88.93 for the term-level task. They implemented various features, including surface-form, semantic, and sentiment features, and generated two large word-sentiment association lexicons: one from tweets with sentiment-word hashtags and another from tweets with emoticons. The lexicons significantly improved performance, particularly in the message-level task, where they contributed a gain of over 5 F-score points. The systems are replicable using freely available resources.This paper describes the development of two state-of-the-art SVM classifiers for sentiment analysis of tweets and SMS messages. The first classifier detects the overall sentiment of a message (message-level task), while the second classifier identifies the sentiment of individual terms within a message (term-level task). The authors participated in the SemEval-2013 competition and achieved the highest scores in both tasks, with an F-score of 69.02 for the message-level task and 88.93 for the term-level task. They implemented various features, including surface-form, semantic, and sentiment features, and generated two large word-sentiment association lexicons: one from tweets with sentiment-word hashtags and another from tweets with emoticons. The lexicons significantly improved performance, particularly in the message-level task, where they contributed a gain of over 5 F-score points. The systems are replicable using freely available resources.