I. Introduction
Understanding information and converting it into a knowledge is a challenging issue in the history of big data. Recently, many studies are raising awareness in natural language processing for Arabic and its dialects. This growth of awareness can be recognized because of the extensive use of Arabic dialects [1]. In concepts of Arabic language dialect processing and specifically in sentiment understanding, this typically involves realizing different human languages which are obliquely related to appropriate human aspects and concepts. A very brief example of actual dissimilarities can be shown by a word mabsut, which has a positive insight in Egypt, however, it does not has the same insight in other Arabic dialects such as Iraq [2]. In this circumstance, the social websites such as Facebook and Twitter have been considered to be the source of useful data for trend analysis, taking into account the volume of the data and its popularity in different dialects. Currently, mining Facebook proved to be an appropriate area for a research, as it is a source of newly daily information [3]. Such enormous amounts of producing data may become useful to businesses, government, and others related creation essential decisions. Competition between companies in the market is an old problem that becomes more challenging by the appearance of social media [4].