04/16/2018

SPIE Disruptive Technologies in Information Science: Reservoir Presentation Topic Modeling for Analysis of Big Data Tensor Decomposition

On Tuesday, April 17 in Orlando, FL, Tom Henretty, Managing Engineer at Reservoir Labs, will present the paper “Topic Modeling for Analysis of Big Data Tensor Decomposition” at the SPIE Disruptive Technologies in Information Science conference. Tensor decompositions have recently shown promising results when addressing problems in cyber security and intelligence analysis. Tensor decomposition can pull out components, kinds of stories, activities, and anomalies in big data.  Reservoir’s ENSIGN technology enables tensor decompositions to scale to very large data. Tom will present an approach to automated component clustering and classification based on the Latent Dirichlet Allocation (LDA) topic modeling technique. This enables more rapid separation of known activities from anomalies. Tom will show example applications to representative cyber security and geospatial datasets. For more information about Reservoir Labs’ presentation, click here.