ASTD: Arabic Sentiment Tweets Dataset

Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 2515–2519, Lisbon, Portugal, 17-21 September 2015. c⃝2015 Association for Computational Linguistics. ASTD: Arabic Sentiment Tweets Dataset Mahmoud Nabil Computer Engineering Cairo University Giza, Egypt mah.nabil@cu.edu.eg Mohamed Aly Computer Engineering Cairo University Giza, Egypt mohamed@mohamedaly.info Amir F. Atiya Computer Engineering Cairo University Giza, Egypt amir@alumni.caltech.edu Abstract This paper introduces ASTD, an Arabic social sentiment analysis dataset gathered from Twitter. It consists of about 10,000 tweets which are classified as objective, subjective positive, subjective negative, and subjective mixed. We present the properties and the statistics of the dataset, and run experiments using standard par- titioning of the dataset. Our experiments provide benchmark results for 4 way sen- timent classification on the dataset. 1 Introduction Arabic sentiment analysis work is gaining large at- tention nowadays. This is mainly due to the need of a product that can utilize natural language pro- cessing technology to track and analyze the public mood through processing social data streams. This calls for using standard social sentiment analysis datasets. In this work we present ASTD (Arabic Sentiment Tweets Dataset) an Arabic social sen- timent analysis dataset gathered from Twitter. We discuss our method for gathering and annotating the dataset, and present its properties and statis- tics through the following tasks: (1) 4 way sen- timent classification (2) Two stage class classifi- cation; and (3) sentiment lexicon generation. The contributions in this work can be summarized as: 1. We present an Arabic social dataset of about 10k tweets for subjectivity and sentiment analysis gathered from. 2. We investigate the properties and the statis- tics of the dataset and provide standard splits for balanced and unbalanced settings of the dataset. 3. We present a set of benchmark experiments to the dataset to establish a baseline for future comparisons. 4. We make the dataset and the used experi- ments publicly available1. 2 Related Work The detection of user sentiment in texts is a re- cent task in natural language processing. This task is gaining a large attention nowadays due to the explosion in the number of social media plat- forms and the number of people using them. Some Arabic sentiment datasets have been collected (see Table 1). (Abdul-Mageed et al., 2014) pro- posed the SAMAR system that perform subjectiv- ity and sentiment analysis for Arabic social media where they used different multi-domain datasets collected from Wikipedia TalkPages, Twitter, and Arabic forums. (Aly and Atiya, 2013) proposed LABR, a book reviews dataset collected from GoodReads. (Rushdi-Saleh et al., 2011) presented an Arabic corpus of 500 movie reviews collected from different web pages. (Refaee and Rieser, 2014) presented a manually annotated Arabic so- cial corpus of 8,868 Tweets and they discussed the method of collecting and annotating the cor- pus. (Abdul-Mageed and Diab, 2014) proposed SANA, a large-scale, multi-domain, and multi- genre Arabic sentiment lexicon. The lexicon au- tomatically extends two manually collected lex- icons HUDA (4,905 entries) and SIFFAT (3,325 entries). (Ibrahim et al., 2015) built a manual cor- …

Content truncated to 500 words