Outlier detection for text data

Ramakrishnan Kannan; Hyenkyun Woo; Charu C. Aggarwal; Haesun Park

doi:10.1137/1.9781611974973.55

Outlier detection for text data

Ramakrishnan Kannan
, Hyenkyun Woo
, Charu C. Aggarwal
, Haesun Park

Learning Systems

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

48 Scopus citations

Abstract

The problem of outlier detection is extremely challenging in many domains such as text, in which the attribute values are typically non-negative, and most values are zero. In such cases, it often becomes difficult to separate the outliers from the natural variations in the patterns in the underlying data. In this paper, we present a matrix factorization method, which is naturally able to distinguish the anomalies with the use of low rank approximations of the underlying data. Our iterative algorithm TONMF is based on Block Coordinate Descent (BCD) framework. Our approach has significant advantages over traditional methods for text outlier detection. Finally, we present experimental results illustrating the effectiveness of our method over competing methods.

Original language	English
Title of host publication	Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017
Editors	Nitesh Chawla, Wei Wang
Publisher	Society for Industrial and Applied Mathematics Publications
Pages	489-497
Number of pages	9
ISBN (Electronic)	9781611974874
DOIs	https://doi.org/10.1137/1.9781611974973.55
State	Published - 2017
Event	17th SIAM International Conference on Data Mining, SDM 2017 - Houston, United States Duration: Apr 27 2017 → Apr 29 2017

Publication series

Name	Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017

Conference

Conference	17th SIAM International Conference on Data Mining, SDM 2017
Country/Territory	United States
City	Houston
Period	04/27/17 → 04/29/17

Funding

This manuscript has been co-authored by UT-Battelle, LLC under Contract No. DE-AC05-00OR22725 with the U.S. Department of Energy. This project was partially funded by the Laboratory Director’s Research and Development fund, National Science Foundation (NSF) grant IIS-1348152, Defense Advanced Research Projects Agency (DARPA) XDATA program grant FA8750-12-2-0309 and also sponsored by the Army Research Laboratory (ARL) accomplished under Cooperative Agreement Number W911NF-09-2-0053. Also, H. Woo is supported by NRF-2015R101A1A01061261.

Access to Document

10.1137/1.9781611974973.55

Cite this

Kannan, R., Woo, H., Aggarwal, C. C., & Park, H. (2017). Outlier detection for text data. In N. Chawla, & W. Wang (Eds.), Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017 (pp. 489-497). (Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017). Society for Industrial and Applied Mathematics Publications. https://doi.org/10.1137/1.9781611974973.55

Kannan, Ramakrishnan ; Woo, Hyenkyun ; Aggarwal, Charu C. et al. / Outlier detection for text data. Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017. editor / Nitesh Chawla ; Wei Wang. Society for Industrial and Applied Mathematics Publications, 2017. pp. 489-497 (Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017).

@inproceedings{20b42fedbbd74086a4eab38f2d1f5cd7,

title = "Outlier detection for text data",

abstract = "The problem of outlier detection is extremely challenging in many domains such as text, in which the attribute values are typically non-negative, and most values are zero. In such cases, it often becomes difficult to separate the outliers from the natural variations in the patterns in the underlying data. In this paper, we present a matrix factorization method, which is naturally able to distinguish the anomalies with the use of low rank approximations of the underlying data. Our iterative algorithm TONMF is based on Block Coordinate Descent (BCD) framework. Our approach has significant advantages over traditional methods for text outlier detection. Finally, we present experimental results illustrating the effectiveness of our method over competing methods.",

author = "Ramakrishnan Kannan and Hyenkyun Woo and Aggarwal, \{Charu C.\} and Haesun Park",

note = "Publisher Copyright: Copyright {\textcopyright} by SIAM.; 17th SIAM International Conference on Data Mining, SDM 2017 ; Conference date: 27-04-2017 Through 29-04-2017",

year = "2017",

doi = "10.1137/1.9781611974973.55",

language = "English",

series = "Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017",

publisher = "Society for Industrial and Applied Mathematics Publications",

pages = "489--497",

editor = "Nitesh Chawla and Wei Wang",

booktitle = "Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017",

}

Kannan, R, Woo, H, Aggarwal, CC & Park, H 2017, Outlier detection for text data. in N Chawla & W Wang (eds), Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017. Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017, Society for Industrial and Applied Mathematics Publications, pp. 489-497, 17th SIAM International Conference on Data Mining, SDM 2017, Houston, United States, 04/27/17. https://doi.org/10.1137/1.9781611974973.55

Outlier detection for text data. / Kannan, Ramakrishnan; Woo, Hyenkyun; Aggarwal, Charu C. et al.
Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017. ed. / Nitesh Chawla; Wei Wang. Society for Industrial and Applied Mathematics Publications, 2017. p. 489-497 (Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Outlier detection for text data

AU - Kannan, Ramakrishnan

AU - Woo, Hyenkyun

AU - Aggarwal, Charu C.

AU - Park, Haesun

PY - 2017

Y1 - 2017

N2 - The problem of outlier detection is extremely challenging in many domains such as text, in which the attribute values are typically non-negative, and most values are zero. In such cases, it often becomes difficult to separate the outliers from the natural variations in the patterns in the underlying data. In this paper, we present a matrix factorization method, which is naturally able to distinguish the anomalies with the use of low rank approximations of the underlying data. Our iterative algorithm TONMF is based on Block Coordinate Descent (BCD) framework. Our approach has significant advantages over traditional methods for text outlier detection. Finally, we present experimental results illustrating the effectiveness of our method over competing methods.

AB - The problem of outlier detection is extremely challenging in many domains such as text, in which the attribute values are typically non-negative, and most values are zero. In such cases, it often becomes difficult to separate the outliers from the natural variations in the patterns in the underlying data. In this paper, we present a matrix factorization method, which is naturally able to distinguish the anomalies with the use of low rank approximations of the underlying data. Our iterative algorithm TONMF is based on Block Coordinate Descent (BCD) framework. Our approach has significant advantages over traditional methods for text outlier detection. Finally, we present experimental results illustrating the effectiveness of our method over competing methods.

UR - https://www.scopus.com/pages/publications/85027865472

U2 - 10.1137/1.9781611974973.55

DO - 10.1137/1.9781611974973.55

M3 - Conference contribution

AN - SCOPUS:85027865472

T3 - Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017

SP - 489

EP - 497

BT - Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017

A2 - Chawla, Nitesh

A2 - Wang, Wei

PB - Society for Industrial and Applied Mathematics Publications

T2 - 17th SIAM International Conference on Data Mining, SDM 2017

Y2 - 27 April 2017 through 29 April 2017

ER -

Kannan R, Woo H, Aggarwal CC, Park H. Outlier detection for text data. In Chawla N, Wang W, editors, Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017. Society for Industrial and Applied Mathematics Publications. 2017. p. 489-497. (Proceedings of the 17th SIAM International Conference on Data Mining, SDM 2017). doi: 10.1137/1.9781611974973.55

Outlier detection for text data

Abstract

Publication series

Conference

Funding

Access to Document

Other files and links

Fingerprint

Cite this