What we've developed is an algorithm for unsupervised extraction of meta-data across blogs independent of language. We've implemented the algorithm as a way of automatically extracting text data in a clean format from blogs. We also have datasets that show a comparison of this algorithm with previous techniques on the same data.