Résumé:
oday, sentiment analysis holds great importance in various fields such as politics,
production, and services, among others. Currently, social media platforms are filled
with texts in which internet users express their opinions on different subjects, and the
value of their opinions is considerable. Understanding the content conveyed by these
texts is essential. It can be said that a good manager is one who attentively listens to
the opinions of citizens. In this regard, sentiment analysis is highly significant in
meeting the needs of citizens.
In this work, we will create an LSTM model that will analyze and classify a set of posts
from social media. Subsequently, we will develop a web application for analyzing user
opinions. The defined classes are positive and negative. This work stands out among
the few that utilize models to analyze comments using the Algerian dialect.
The obtained results are encouraging, with an accuracy of 84.7%, demonstrating the
system’s ability to distinguish between negative and positive discourse.
