A method for probabilistically finding the connected components of an undirected graph. The method includes identifying a first edge, having a first and second vertex, and inserting information detailing the first and second vertex of the first edge into a bloom filter associated with a root node of a bloom filter data structure. A first node, connected to the root node, is created, comprising an associated bloom filter containing information associated with the first and second vertex of the first edge. The method includes identifying a second edge, having a first and second vertex, and inserting information detailing the first and second vertex of the second edge into a bloom filter associated with the root node of the bloom filter data structure. A second node, connected to the root node, is created, comprising an associated bloom filter containing information associated with the first and second vertex of the second edge.