This thesis introduces a novel approach for identifying essential genes and cancer-related modules using a network-based unsupervised learning model, coupled with a Python toolkit that translates these findings into practical applications. The model constructs a gene network from frequently mutated genes and their associated biological processes, computes topological features to rank these genes, and employs a modified greedy algorithm to detect critical cancer-related modules. The evaluation of this approach across multiple cancer types demonstrates its efficacy in prioritizing genes with known roles in cancer and uncovering biologically relevant modules. To facilitate the application of these findings, a Python-based toolkit was developed, enabling researchers and clinicians to input gene lists, obtain rankings based on cancer relevance, and identify associated gene modules. This toolkit serves as an accessible resource for genomic analysis, advancing the field of cancer genomics by providing a practical tool for identifying potential therapeutic targets. This work contributes to the advancement of precision oncology by offering a robust method for gene prioritization and module detection, along with a user-friendly toolkit that enhances research and clinical decision-making in cancer genomics.