ABSTRACT The rapid growth of social media platforms has generated a vast amount of user-generated textual data. These platforms contain valuable information reflecting public opinions, emotions, and attitudes toward products, events, and services. Sentiment analysis, also known as opinion mining, is an important application of natural language processing that aims to identify the emotional tone behind textual content. Traditional machine learning methods have been widely used for sentiment classification; however, they often struggle to capture contextual relationships within text sequences. Deep learning models, particularly Long Short-Term Memory (LSTM) networks, have shown promising performance in analyzing sequential textual data. This study proposes a deep learning-based sentiment analysis framework using LSTM for classifying social media posts into positive, negative, and neutral sentiments. Text data is preprocessed and transformed into numerical representations using word embeddings. The performance of the proposed model is evaluated using Accuracy, Precision, Recall, and F1-Score metrics. Experimental results indicate that LSTM networks effectively capture contextual dependencies in text data and provide reliable sentiment classification performance. Key words: Sentiment Analysis, Deep Learning, LSTM, Natural Language Processing, Social Media Analytics