In the paper an adapted version of the leaders clustering method as an efficient method for clustering ego-centered networks is presented. The original data are transformed into symbolic objects. Clustering problem is defined as an optimization problem. Based on this definition an adapted leaders method was developed. The example on the dataset on social support in Ljubljana collected by the Faculty of social sciences at the University of Ljubljana is presented. The clustering of the data was done using the adapted leaders program from the program package CLAMIX.