A telecommunication network produces daily large amounts of calling data which contain hidden and valuable knowledge. This knowledge can be used in determining the calling patterns of customer, finding too thrifty customers, also for locating the best area to concentrate on in order to boost profits. In this paper, we designed an algorithm named CLOTELE which is based on the ideas of the pattern-growth method of mining, for mining closed frequent calling patterns of a telecommunication database from a telecommunication provider. First, by observing the features of the database and then extracting the attributes needed to be mined. Then, we merge the items to form an itemset, the algorithm is now applied to the transformed database. CLOTELE is built on the basis of viewing knowledge discovery as an interactive and iterative process in order to optimize decision making. The quality of the knowledge discovered is evaluated. The experimental results show the knowledge of closed frequent patterns obtained is very useful and easy to interpret by the network operators and administrators.