Ontologies are domain-specific constructs developed for many different purposes. But, they can be different even within the same domain. Ontology matching is a method for finding the same things in between existing ontologies by looking at semantic similarities. Especially, ontology matching algorithms are used for data integration, reuse of existing ontologies, and information discovery. However, the matching process is quite difficult to perform for large ontologies. For this reason, it is recommended that ontologies can be divided into small pieces and matched modularly. We aim to automatically separate ontologies into sub-segments in this scope of the study. Primarily, the ontology data sets have been converted into a graph structure. Then, We test two algorithms to partition graph. The first algorithm is Karger Algorithm, which finds min-cut edge set in the graph, and the second algorithm is based clique percolation. The test results for both algorithms are shown and the advantages and disadvantages of these two algorithms are explained.