Data set Description.txt

1. Title: Twitter Data set for Arabic Sentiment Analysis

2. Sources:
     (a) Creator: N. A. Abdulla and N. Mahyoub
     (b) Donor: N. A. Abdulla (naabdulla11@cit.just.edu.jo)
     (c) Date: April, 2014

3. Past Usage:
   - Publications:
   1.  Abdulla N. A., Mahyoub N. A., Shehab M., Al-Ayyoub M.,
        “Arabic Sentiment Analysis: Corpus-based and Lexicon-based”,
        IEEE conference on Applied Electrical Engineering and Computing Technologies (AEECT 2013),
        December 3-12, 2013, Amman, Jordan. (Accepted for Publication).

4. Relevant Information:
   --- By using a tweet crawler, we collect 2000 labelled tweets (1000 positive tweets and 1000 negative ones)
       on various topics such as: politics and arts. These tweets include opinions written in both
       Modern Standard Arabic (MSA) and the Jordanian dialect.

   --- The selected tweets convey some kind of feelings (positive or negative) and the objective of our model is
        to extract valuable information from such tweets in order to determine the sentiment orientation of the inputted text.
       The months-long annotation process of the tweets is manually conducted mainly by two human experts
       (native speakers of Arabic). If both experts agree on the label of a certain tweet, then the tweet is assigned this label.
       Otherwise, a third expert is consulted to break the tie. 

   --- Predicted attribute: class of opinion polarity.
    

5. Number of Instances: 2000 (1000 in each of two classes)

6. Number of Attributes: string attribute (vector of words) and the class

7. Attribute Information:
   1. Tweet as a string vector
   2. class: 
      -- Positive polarity
      -- Negative poalrity

8. Missing Attribute Values: None

Summary Statistics:
	 			Positive	Negative        
  	Total tweets		1000	1000
	Total words		7189	9769
	Avg. words in each tweet	7.19	9.97
	Avg. characters in each tweet	40.04	59.02


9. Class Distribution: 50% for each of 2 classes.