This algorithm measures correlations in large volume data streams quickly and efficiently. Large data streams are typically only read once because storage would be computationally burdensome and expensive. Keeping meaningful statistics to summarize a data stream, has therefore become important but calculating correlations between data elements in this way is particularly challenging. This algorithm uses a novel combinatorial approach so that correlations between multidimensional data can readily be acquired, while maintaining efficient dimensionality reduction. Analysis of internet traffic is one important application area.