This article is intended for the hybrid deep learning model specifically designed to effectively summarize social media posts with the exponential growth of social media, extracting meaningful insights from the vast amount of user-generated content has become a significant challenge. This paper presents a hybrid deep learning model specifically designed to effectively summarize social media posts. It combines extractive summarization, which selects key information, and abstractive summarization, which rephrases content into shorter, coherent summaries. This dual approach is particularly suited to handle the informal, concise, and diverse language typical of social media. It employs a Transformer-based architecture to generate summaries and utilizes a graph-based attention mechanism to identify and prioritize important relationships between posts and comments. To address the rapidly changing nature of social media language, advanced preprocessing techniques are implemented, including sentiment analysis, hashtag normalization, and the resolution of ambiguous references. With the explosion of social media content, extracting valuable insights from vast user-generated data has become increasingly difficult. This paper introduces a hybrid deep learning model tailored for summarizing social media posts. It merges extractive summarization, which highlights key points, with abstractive summarization, which rewrites the content into concise summaries. The model uses a Transformer-based architecture and graph-based attention mechanism to focus on important relationships between posts and comments. To handle the constantly evolving language of social media, advanced preprocessing steps like sentiment analysis, hashtag normalization, and ambiguity resolution are incorporated. A Reinforcement Learning (RL) component fine-tunes the summaries for improved relevance, fluency, and informativeness. The proposed model surpasses existing methods in performance metrics like ROUGE and BLEU, proving its effectiveness in summarizing content filled with slang, abbreviations, and emerging trends. This solution is particularly useful for applications such as news aggregation, sentiment analysis, and customer feedback summarization, offering a robust approach to managing large volumes of unstructured social media data.