Research proposal classification is an important task for private research funding agencies. When a large number of research proposals are received, it is common to group them according to their similarities in research disciplines. Current methods for grouping proposals are based on manual matching of similar research discipline areas and/or keywords. This paper presents a pattern-based text-mining approach to classify research proposals based on their similarities in research areas. This paper focus on developing efficient mining algorithm for discovering patterns from large data collection and search for useful and interesting patterns.