Spark at Akanoo

Most (user-generated) data of the internet, but also genomics data, weather data, business or industrial process data etc. is sequential in nature. Finding frequent sequential patterns is key to many business and research applications. The PrefixSpan algorithm is an efficient way to calculate frequent patterns from huge sets of patterns. Akanoo has implemented the PrefixSpan algorithm in Scala using Apache Spark.

Presented by:
Frank Wolf
Dr. Gundula Meckenhäuser