Literature review of Social media platforms

        Sentiment analysis or opinion mining has gained much attention in recent years; therefore, it is important to review recent resources for sentiment analysis to gain meaningful knowledge.

        In the work presented in [8] Kursuncu, Gaur, Lokala, Thirunarayan, Sheth and Arpinar they provide details about the applications of sentiment analysis in different domains such as health care, political and social issues, disaster management, sales and stock predictions. In the domain of the social issues, they present that social issues and related events have been a part of discussions on Twitter, which gives opportunities to researchers in addressing problems concerning individuals as well as the society at large. Solutions to such problems can be provided by measuring public opinions on Twitter by employing predictive analysis.

        In [9] YU and Albaadani introduced a new sentiment classification scale. They classified sentiments as Highly Positive (HP), Fairly Positive (FP), No Sentiment (NS), Fairly Negative (FN) and Highly Negative (HN). They presented a comprehensive approach to Arabic social media sentiment analysis by combining lexicon-based ideas with machine learning techniques. Anastasia and Fabio identified some of sentiment analysis challenges in their study [10].One of the challenges is tweets that contain a multimedia content such as Image or video because it is difficult to extract the information from multimedia tweets. Finally, some tweets are written in mixed languages which make it difficult to detect and analyze the tweet. Alhumoud, Altuwaijri, Albuhairi, and Alohaideb in [11] showed other challenges of analyzing Arabic text such as that every part of Saudi Arabia has it is own version or dialect of Arabic. Also, some Arabic words could have the same word-spelling, but with a different meaning depending on its punctuation.

        Another aspect that is worth mentioning by Alowisheq, Alhumoud, Altwairesh, and Albuhairi in [12] says that most of sentiment analysis works focus on English language and there is relatively less work on Arabic language compared to English. The complexity of Arabic language and the lack of the available resources of Arabic sentiment analysis like lexicons and datasets are the main obstacles in Arabic sentiment analysis.

    Mishra, Rajnish and Kumar in [13] discussed the steps used to perform sentiment analysis starting with data collection, pre-processing data, feature extraction, sentiment analysis through dictionary matching and polarity classification. AlMurtadha in [14] says that analyzing recent Twitter hash tags trends helps tobetter understand the public opinions about any topic. Also, the result of his study shows that the more tweets retrieved from a hashtag, the more an accurate the result will be.

Sentiments i[u1] s a mobile application developed by Ziya Bal, it  requires iOS 7.0 or later. Compatible with iPhone, iPad, and iPod touch. It discovers how people feel about a particular topic. The application determines if the people’s opinion or attitude is negative, positive or neutral this process known as opinion mining. The user types an English  word or sentence in the input field then the application will present the result as a percentage, the background color will be changed accordingly, if it’s positive the background will be green or red if it’s negative [15].