This paper describes our sentiment analysis systems which have been built for SemEval-2015 Task 10 Subtask B and E. For subtask B, a Logistic Regression classifier has been trained after extracting several groups of features including lexical, syntactic, lexiconbased, Z score and semantic features. A weighting schema has been adapted for positive and negative labels in order to take into account the unbalanced distribution of tweets between the positive and negative classes. This system is ranked third over 40 participants, it achieves average F1 64.27 on Twitter data set 2015 just 0.57% less than the first system. We also present our participation in Subtask E in which our system has got the second rank with Kendall metric but the first one with Spearman for ranking twitter terms according to their association with the positive sentiment.

This paper describes our contribution in Opinion Target Extraction OTE and Sentiment Polarity sub tasks of SemEval 2015 ABSA task. A CRF model with IOB notation has been adopted for OTE with several groups of features including syntactic, lexical, semantic, sentiment lexicon features. Our submission for OTE is ranked fifth over twenty submissions. A Logistic Regression model with a weighting schema of positive and negative labels have been used for sentiment polarity; several groups of features (lexical, syntactic, semantic, lexicon and Z score) are extracted. Our submission for Sentiment Polarity is ranked third over ten submissions on the restaurant data set, third over thirteen on the laptops data set, but the first over eleven on the hotel data set that is out-of-domain set.

Sentiment lexicon-based features have proved their performance in recent work concerning sentiment analysis in Twitter. Automatic constructed lexicon features seem to be enough influential to attract the attention. In this paper, we propose a new metric to estimate the word polarity score, called natural entropy (ne), in order to construct a new sentiment lexicon based on Sentiment140 corpus. We derive six features from the new lexicon and show that (ne) metric outperforms the PMI metric which has been used for the same purpose. For evaluation, we build a state-of-the-art system for sentiment analysis in short text using a supervised classifier trained on several groups of features including n-gram, sentiment lexicons, negation, Z score and semantic features. This system has been one of the best systems in both tasks of SemEval-2015: Sentiment Analysis in Twitter and Aspect-Based Sentiment Analysis. We investigate the impact of the lexicon-based features extracted from existing manual and automatic constructed lexicons on the system performance and also the impact of the proposed metric (ne).

In this paper, we present our contribution in SemEval2014 ABSA task, some supervised methods for Aspect-Based Sentiment Analysis of restaurant and laptop reviews are proposed, implemented and evaluated. We focus on determining the aspect terms existing in each sentence, finding out their polarities, detecting the categories of the sentence and the polarity of each category. The evaluation results of our proposed methods exhibit a significant improvement in terms of accuracy and f-measure over all four subtasks regarding to the baseline proposed by SemEval organisers.

Twitter has become more and more an important resource of user-generated data. Sentiment Analysis in Twitter is interesting for many applications and objectives. In this paper, we propose to exploit some features which can be useful for this task; the main contribution is the use of Z-scores as features for sentiment classification in addition to pre-polarity and POS tags features. Our experiments have been evaluated using the test data provided by SemEval 2013 and 2014. The evaluation demonstrates that Z_scores features can significantly improve the prediction performance.