With so much information available online, it’s more important than ever to have reliable tools for summarizing text quickly and accurately. In this paper, we introduce a new way to improve the popular TextRank algorithm for extractive summarization. By adding a dynamic damping factor and using Latent Dirichlet Allocation (LDA) to enhance how text is represented, our method creates more meaningful summaries. We tested it with metrics like Pyramid, METEOR, and ROUGE, and compared it to the original TextRank. The results were promising, showing that our approach produces better summaries and could be useful for real-world applications like text mining and information retrieval.