STExNMF: Spatio-Temporally Exclusive Topic Discovery for Anomalous Event Detection

Shin, Sungbok; Choi, Minsuk; Choi, Jinho; Langevin, Scott; Bethune, Christopher; Horne, Philippe; Kronenfeld, Nathan; Kannan, Ramakrishnan {ramki}; Drake, Barry; Park, Haesun

doi:10.1109/ICDM.2017.53

Title: STExNMF: Spatio-Temporally Exclusive Topic Discovery for Anomalous Event Detection

Full Record
Other Related Research

Abstract

Understanding newly emerging events or topics associated with a particular region of a given day can provide deep insight on the critical events occurring in highly evolving metropolitan cities. We propose herein a novel topic modeling approach on text documents with spatio-temporal information (e.g., when and where a document was published) such as location-based social media data to discover prevalent topics or newly emerging events with respect to an area and a time point. We consider a map view composed of regular grids or tiles with each showing topic keywords from documents of the corresponding region. To this end, we present a tilebased spatio-temporally exclusive topic modeling approach called STExNMF, based on a novel nonnegative matrix factorization (NMF) technique. STExNMF mainly works based on the two following stages: (1) first running a standard NMF of each tile to obtain general topics of the tile and (2) running a spatiotemporally exclusive NMF on a weighted residual matrix. These topics likely reveal information on newly emerging events or topics of interest within a region. We demonstrate the advantages of our approach using the geo-tagged Twitter data of New York City. We also provide quantitative comparisons in terms of the topic quality, spatio-temporalmore »« less

Authors:

Shin, Sungbok ^[1]; Choi, Minsuk ^[1]; Choi, Jinho ^[1]; Langevin, Scott ^[2]; Bethune, Christopher ^[2]; Horne, Philippe ^[2]; Kronenfeld, Nathan ^[2];

^[3]; Drake, Barry ^[4]; Park, Haesun ^[5]

Korea University
Uncharted Software Inc
ORNL
Georgia Tech Research Institute (GTRI), Atlanta, GA
Georgia Institute of Technology, Atlanta

Publication Date:: 2017-11-01

Research Org.:: Oak Ridge National Laboratory (ORNL), Oak Ridge, TN (United States)

Sponsoring Org.:: USDOE

OSTI Identifier:: 1426578

DOE Contract Number:: AC05-00OR22725

Resource Type:: Conference

Resource Relation:: Conference: IEEE International Conference on Data Mining (ICDM) - New Orleans, Louisiana, United States of America - 11/18/2017 3:00:00 PM-11/21/2017 3:00:00 PM

Country of Publication:: United States

Language:: English

Citation Formats


                    Shin, Sungbok, Choi, Minsuk, Choi, Jinho, Langevin, Scott, Bethune, Christopher, Horne, Philippe, Kronenfeld, Nathan, Kannan, Ramakrishnan, Drake, Barry, and Park, Haesun. STExNMF: Spatio-Temporally Exclusive Topic Discovery for Anomalous Event Detection.  United States: N. p., 2017. 
        Web.  doi:10.1109/ICDM.2017.53.

Copy to clipboard


                    Shin, Sungbok, Choi, Minsuk, Choi, Jinho, Langevin, Scott, Bethune, Christopher, Horne, Philippe, Kronenfeld, Nathan, Kannan, Ramakrishnan, Drake, Barry, & Park, Haesun. STExNMF: Spatio-Temporally Exclusive Topic Discovery for Anomalous Event Detection.  United States.  https://doi.org/10.1109/ICDM.2017.53

Copy to clipboard


                    Shin, Sungbok, Choi, Minsuk, Choi, Jinho, Langevin, Scott, Bethune, Christopher, Horne, Philippe, Kronenfeld, Nathan, Kannan, Ramakrishnan, Drake, Barry, and Park, Haesun. 2017.  
        "STExNMF: Spatio-Temporally Exclusive Topic Discovery for Anomalous Event Detection".  United States.  https://doi.org/10.1109/ICDM.2017.53.  https://www.osti.gov/servlets/purl/1426578.

Copy to clipboard


                    
@article{osti_1426578,

  title        = {STExNMF: Spatio-Temporally Exclusive Topic Discovery for Anomalous Event Detection},

  author       = {Shin, Sungbok and Choi, Minsuk and Choi, Jinho and Langevin, Scott and Bethune, Christopher and Horne, Philippe and Kronenfeld, Nathan and Kannan, Ramakrishnan and Drake, Barry and Park, Haesun},

  abstractNote = {Understanding newly emerging events or topics associated with a particular region of a given day can provide deep insight on the critical events occurring in highly evolving metropolitan cities. We propose herein a novel topic modeling approach on text documents with spatio-temporal information (e.g., when and where a document was published) such as location-based social media data to discover prevalent topics or newly emerging events with respect to an area and a time point. We consider a map view composed of regular grids or tiles with each showing topic keywords from documents of the corresponding region. To this end, we present a tilebased spatio-temporally exclusive topic modeling approach called STExNMF, based on a novel nonnegative matrix factorization (NMF) technique. STExNMF mainly works based on the two following stages: (1) first running a standard NMF of each tile to obtain general topics of the tile and (2) running a spatiotemporally exclusive NMF on a weighted residual matrix. These topics likely reveal information on newly emerging events or topics of interest within a region. We demonstrate the advantages of our approach using the geo-tagged Twitter data of New York City. We also provide quantitative comparisons in terms of the topic quality, spatio-temporal exclusiveness, topic variation, and qualitative evaluations of our method using several usage scenarios. In addition, we present a fast topic modeling technique of our model by leveraging parallel computing.},

  doi          = {10.1109/ICDM.2017.53},

  url          = {https://www.osti.gov/biblio/1426578},
  journal      = {},
number       = ,

  volume       = ,

  place        = {United States},

  year         = {2017},

  month        = {11}

}

Copy to clipboard

Conference:

View Conference (5.30 MB)

https://doi.org/10.1109/ICDM.2017.53

Other availability

Please see Document Availability for additional information on obtaining the full-text document. Library patrons may search WorldCat to identify libraries that hold this conference proceeding.

Save / Share:

Export Metadata

Save to My Library

Similar records in OSTI.GOV collections:

TopicOnTiles: Tile-Based Spatio-Temporal Event Analytics via Exclusive Topic Modeling on Social Media

Conference Choi, Minsuk; Shin, Sungbok; Choi, Jinho; ...

Detecting anomalous events of a particular area in a timely manner is an important task. Geo-tagged social media data are useful resource for this task; however, the abundance of everyday language in them makes this task still challenging. To address such challenges, we present TopicOnTiles, a visual analytics system that can reveal information relevant to anomalous events in a multi-level tile-based map interface by using social media data. To this end, we adopt and improve a recently proposed topic modeling method that can extract spatio-temporally exclusive topics corresponding to a particular region and a time point. Furthermore, we utilize amore »« less
https://doi.org/10.1145/3173574.3174157

Full Text Available
Connect the Dots: In Situ 4-D Seismic Monitoring of CO₂ Storage With Spatio-Temporal CNNs

Journal Article Feng, Shihang; Zhang, Xitong; Wohlberg, Brendt; ... - IEEE Transactions on Geoscience and Remote Sensing

4-D seismic imaging has been widely used in CO₂ sequestration projects to monitor the fluid flow in the volumetric subsurface region that is not sampled by wells. Ideally, real-time monitoring and near-future forecasting would provide site operators with great insights to understand the dynamics of the subsurface reservoir and assess any potential risks. However, due to obstacles such as high deployment cost, availability of acquisition equipment, exclusion zones around surface structures, only very sparse seismic imaging data can be obtained during monitoring. That leads to an unavoidable and growing knowledge gap over time. The operator needs to understand the fluidmore »« less
https://doi.org/10.1109/tgrs.2021.3116618

Full Text Available
A Novel Spatio-Temporal Regime Tracking Method for Impact Simulations

Technical Report Harwell, Megan

In this proposal, we present a novel method of tracking the rheological regimes activated during impact cratering events that will allow researchers to gain new insights into cratering mechanics. Rheology describes the stress-strain response of rocks to different conditions. Planetary impact cratering events often occur on too large of a scale to be feasibly captured in controlled experiments. Instead, these dynamic events are primarily studied using multi-physics codes equipped with complex material models that enable calculations of the impact event at scale. However, determining which physical processes are required for the problem of interest is challenging. Because the dominant rheologicalmore »« less
https://doi.org/10.2172/1998099

Full Text Available
An Approximate Entropy Based Approach for Quantifying Stability in Spatio-Temporal Data with Limited Temporal Observations

Conference Piburn, Jesse; Stewart, Robert; Morton, April

Identifying erratic or unstable time-series is an area of interest to many fields. Recently, there have been successful developments towards this goal. These newly developed methodologies however come from domains where it is typical to have several thousand or more temporal observations. This creates a challenge when attempting to apply these methodologies to time-series with much fewer temporal observations such as for spatio-temporal socio-cultural understanding, a domain where a typical time series of interest might only consist of 20-30 annual observations. Identifying instability in spatio-temporal trends is critical for understanding global dynamics and finding areas of potential concern or intervention.more »« less
Full Text Available
Low-rank matrix decomposition and spatio-temporal sparse recovery for STAP radar

Journal Article Sen, Satyabrata - IEEE Journal of Selected Topics in Signal Processing

We develop space-time adaptive processing (STAP) methods by leveraging the advantages of sparse signal processing techniques in order to detect a slowly-moving target. We observe that the inherent sparse characteristics of a STAP problem can be formulated as the low-rankness of clutter covariance matrix when compared to the total adaptive degrees-of-freedom, and also as the sparse interference spectrum on the spatio-temporal domain. By exploiting these sparse properties, we propose two approaches for estimating the interference covariance matrix. In the first approach, we consider a constrained matrix rank minimization problem (RMP) to decompose the sample covariance matrix into a low-rank positivemore »« less
Cited by 44
https://doi.org/10.1109/JSTSP.2015.2464187

Full Text Available

Similar Records