Abstract
With today's growing technology, Internet usage has increased greatly, especially in Saudi Arabia. Saudi people use the Internet and social media, particularly Twitter, as an expressive platform to share their feelings and opinions. Although many studies have analyzed people's sentiment using Twitter, there is still a shortage of applications in analyzing Arabic tweets. In particular, there is an absence of informative tools that can show how people in Saudi Arabia are feeling according to their tweets. This brings to light the idea of "Saudi Mood", which is a web-based application aims at providing a real-time visualization of people emotions from different Saudi's cities based on mining their tweets' text and emojis used taking into consideration the different dialects. Our solution applies a set of language-processing techniques in order to preprocess the tweets. A dataset containing more than 4000 emotional words has been developed and used to classify the tweets into the relevant emotion (happy, sad, angry, scared, surprised). Then, common emotions will be calculated for each city, and the results will be displayed on a dynamic, colorful map of Saudi Arabia in which colors of cities are changing according to the dominant emotion. Also, the website will present the associated trending hashtags for each city along with some interesting emotion statistics. Developing this solution will bring a live informative tool of Saudi Arabia presented in an easy and efficient manner and it is expected to have a positive impact on several aspects.