Approximate query processing has emerged as an approach to dealing with thehuge data volume and complex queries in the environment of data warehouse. In this paper,we present a novel method that provides approximate answers to OLAP queries. Our methodis based on building a compressed (approximate) data cube by a clustering technique and usingthis compressed data cube to provide answers to queries directly, so it improves the performanceof the queries. We also provide the algorithm of the OLAP queries and the confidence intervalsof query results. An extensive experimental study with the OLAP council benchmark showsthe effectiveness and scalability of our cluster-based approach compared to sampling.