Effective text data management is essential in the age of abundant information. In order to improve text document grouping, this study offers a novel hybrid approach that combines discrete differential evolution (DDE) with genetic algorithm (GA). Cluster centroids are started by GA, and then refined by DDE, achieving a balance between global exploration and local optimization. We investigate the application of TF-IDF and Jaccard similarity measures and contrast our approach with industry standard practices. The higher Fowkes–Mallows index and Silhouette score values in the results demonstrate superior clustering performance. In our data-rich era, processing complicated, unstructured text data requires creative solutions. This paper advances the field of text document clustering by providing those.