This paper adopts the idea of nearest neighbor and proposes a new approach called fast intuitive clustering approach (FICA). Besides, FICA also adds the concept of data compression to lower the operating times and coordinates with parameters to reach global search. A series of experiments have been conducted on FICA and other clustering algorithms, like K-means and DBSCAN. According to the simulation results, it is observed that the proposed FICA clustering algorithm outperforms K-means and DBSCAN. FICA can not only to perform good efficiency and correctness but also be applied in large number of data sets. Finally, the proposed FICA is applied in face recognition problem.