9 May 2018 Topic modeling for analysis of big data tensor decompositions
Author Affiliations +
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.
Conference Presentation
© (2018) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Thomas S. Henretty, Thomas S. Henretty, M. Harper Langston, M. Harper Langston, Muthu Baskaran, Muthu Baskaran, James Ezick, James Ezick, Richard Lethin, Richard Lethin, "Topic modeling for analysis of big data tensor decompositions", Proc. SPIE 10652, Disruptive Technologies in Information Sciences, 1065208 (9 May 2018); doi: 10.1117/12.2306933; https://doi.org/10.1117/12.2306933

Back to Top