• DocumentCode
    719466
  • Title

    Investigate the Context Usage of Arabic Proverbs in Twitter

  • Author

    Al-Wehaibi, Rehab Nasser ; Khan, Muhammad Badruddin

  • Author_Institution
    Coll. of Comput. & Inf. Sci., Al-Imam Univ. Riyadh, Riyadh, Saudi Arabia
  • fYear
    2015
  • fDate
    26-29 April 2015
  • Firstpage
    1
  • Lastpage
    8
  • Abstract
    Current technology facilitates and increases connections through social media, allowing individuals everywhere to spread their ideas to the world. One social media platform is Twitter. One characteristic of a tweet is the requirement of conveying a message in a limited number of words. Proverbs are a feature of language that convey messages effectively in the least number of words. Therefore, we selected the analysis of proverb usage in Arabic tweets. We collected and manually classified tweets with Arabic proverbs in order to develop a model, using data and text mining techniques, to understand the context of their usage in the tweets. Each tweet has two types of features: the external feature refers to the structured data behind tweets, and internal features refer to unstructured data in the tweet content, the "text of the tweet itself." According to the experimental results of this study, we found that internal features are best for assessing the context of Arabic proverb usage in tweets. The study is significant in that it can help understand the psychology of the Arab masses using and spreading their messages and ideas in the Twitter ecosystem.
  • Keywords
    data mining; psychology; social networking (online); text analysis; Arab masses psychology; Arabic proverbs; Arabic tweets; Twitter; data mining techniques; social media platform; structured data; text mining techniques; tweet content; Accuracy; Context; Data models; Niobium; Text categorization; Text mining; Twitter;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Cloud Computing (ICCC), 2015 International Conference on
  • Conference_Location
    Riyadh
  • Print_ISBN
    978-1-4673-6617-5
  • Type

    conf

  • DOI
    10.1109/CLOUDCOMP.2015.7149645
  • Filename
    7149645