In recent years, cluster analysis and association analysis have attracted a lot of attention for large data analysis such as biomedical data analysis. This paper proposes a novel algorithm of frequent closed itemset mining. The algorithm addresses two challenges of data mining: mining large and high dimensional data and interpreting the results of data mining. Frequent itemset mining is the key task of association analysis. The algorithm is based on concept lattice structure so that frequent closed itemsets can be generated to reduce the complicity of mining all frequent itemsets and each frequent closed itemset has more information to facilitate interpretation of mining results. From this feature, the paper also discusses the extension of the algorithm for cluster analysis. The experimental results show the efficiency of this algorithm.