Software repositories contain wealth of information about software code, designs, execution history, code and design changes, bug database, software release and software evolution. To meet increased pressure of releasing updated or new versions of software systems due to changing requirements of stakeholder, software are rarely built from scratch. Software reusability is a primary attribute of software quality which aims to create new software systems with a likelihood of using existing software components to add, modify or delete functionalities in order to adapt to new requirements imposed by stakeholders. Software reuse using software components or modules provide a vehicle for planning and re-using already built software components efficiently. In this paper, we propose a framework for our approach to predict software reusable components from existing software repository on the basis of (1) stakeholders intention (requirement) match and (2) similarity index count for better reuse prediction. To effectively manage storage and retrieval of relevant information we use concept of situational method engineering to match and analyze the information for reuse. We use Genetic algorithm, Rabin Karp algorithm for feature selection and classification and k-means clustering methods of data mining to refine our results of prediction in order to better manage and produce high quality software systems within estimated time and cost.