6,000 tweets were downloaded from Twitter using Tweepy in three batches of 2,000 that were found using three keywords in Egyptian Arabic:
- Man/guy - ragil - راجل
- Yes - aywa -أيوا
- Well - kuweiss كويس
In this notebook, we clean Arabic tweets, return ngrams, term frequncies and Tfidf, segment common words found together, and cluster words by topic. Tools: NLTK, googletrans, sklearn feature extraction for text, kmeans clustering, and matplotlib 3D axes.