An Effective Hybrid Stochastic Gradient Descent Arabic Sentiment Analysis with Partial-Order Microwords and Piecewise Differentiation

Abstract

Instagram, Facebook, and Twitter, among other online platforms, have become an inescapable part of our daily lives. These social media platforms are capable of exchanging news, photographs, and other contents. The sentiment analysis on these online data has risen in popularity recently, particularly in Arabic. Unusual language, which differs from the typical format of the language, distinguishes social networking platforms. As a consequence, efficient methods for assessing the vast number of new word permutations that occur on a regular basis in the digital and online environment are required. This paper presents a sentiment classification model relying on microwords and Stochastic Gradient Descent (SGD). Different effectiveness evaluation measures are used to estimate the performance of the suggested model. The suggested method effectively classifies the verification and testing tweets collection with an accuracy of equal to 88.48 percent, as per the simulation findings.