Network motife, patterns of local interconnections with potential functional properties, are important for the analysis of biological networks. To analyse motifs in networks the first step is to find patterns of interest. This paper presents 1) three different concepts for the determination of pattern frequency and 2) an algorithm to compute these frequencies. The different concepts of pattern frequency depend on the reuse of network elements. The presented algorithm finds all or highly frequent patterns under consideration of these concepts. The utility of this method is demonstrated by applying it to biological data.