Share Email Print

Proceedings Paper

Topic modeling for analysis of big data tensor decompositions
Author(s): Thomas S. Henretty; M. Harper Langston; Muthu Baskaran; James Ezick; Richard Lethin
Format Member Price Non-Member Price
PDF $17.00 $21.00

Paper Abstract

Tensor decompositions are a class of algorithms used for unsupervised pattern discovery. Structured, multidimensional datasets are encoded as tensors and decomposed into discrete, coherent patterns captured as weighted collections of high-dimensional vectors known as components. Tensor decompositions have recently shown promising results when addressing problems related to data comprehension and anomaly discovery in cybersecurity and intelligence analysis. However, analysis of Big Data tensor decompositions is currently a critical bottleneck owing to the volume and variety of unlabeled patterns that are produced. We present an approach to automated component clustering and classification based on the Latent Dirichlet Allocation (LDA) topic modeling technique and show example applications to representative cybersecurity and geospatial datasets.

Paper Details

Date Published: 9 May 2018
PDF: 13 pages
Proc. SPIE 10652, Disruptive Technologies in Information Sciences, 1065208 (9 May 2018); doi: 10.1117/12.2306933
Show Author Affiliations
Thomas S. Henretty, Reservoir Labs, Inc. (United States)
M. Harper Langston, Reservoir Labs, Inc. (United States)
Muthu Baskaran, Reservoir Labs, Inc. (United States)
James Ezick, Reservoir Labs, Inc. (United States)
Richard Lethin, Reservoir Labs, Inc. (United States)

Published in SPIE Proceedings Vol. 10652:
Disruptive Technologies in Information Sciences
Misty Blowers; Russell D. Hall; Venkateswara R. Dasari, Editor(s)

© SPIE. Terms of Use
Back to Top
Sign in to read the full article
Create a free SPIE account to get access to
premium articles and original research
Forgot your username?