Combining Burst Detection for Hot Topic Extraction

Hui Fang Ma; Hui Li Ma

doi:10.4028/www.scientific.net/AMR.268-270.1283

Paper Titles

Study on the Navigation of Patrol Robot of Transformer Substation Based on Color Vision and RFID
p.1259

The Development Mode of Low-Carbon Economy: A Carbon Fund Perspective
p.1265

A Resolution Design of the Energy Conservation Based on Energy Chip CS5460A in Home
p.1271

Development and Application of GPS Navigator in Ship
p.1275

Combining Burst Detection for Hot Topic Extraction
p.1283

Applying Data Conversion between Excel and Database in .NET-Based Applications
p.1289

The Influence Factors Analysis of Initial Excess Pore Water Pressure Caused by Construction of Shield Tunnels
p.1295

Research of Modeling Moving Objects Database over Space-Time Grid
p.1301

Research on Measurement Technique for Evaluating Adaptability of Aspect-Oriented Software Architecture
p.1307

HomeAdvanced Materials ResearchAdvanced Materials Research Vols. 268-270Combining Burst Detection for Hot Topic Extraction

Combining Burst Detection for Hot Topic Extraction

Abstract:

As traditional text representations are not suitable for online dynamic streams, this paper presents a hot topic extraction technique that can be used for tracking news topics over time. The model combines individual word burst into the document-word vector representation, which can emphasize the temporally features of text streams. An energy ratio threshold based burst detection approach is proposed and TF-PDF is then combined to weigh the terms. Experiment results demonstrate that this model is effective in topic extraction for news stream and it can better improve the clustering performance.

You might also be interested in these eBooks

View Preview

Info:

Periodical:

Advanced Materials Research (Volumes 268-270)

Pages:

1283-1288

DOI:

https://doi.org/10.4028/www.scientific.net/AMR.268-270.1283

Citation:

Cite this paper

Online since:

July 2011

Authors:

Hui Fang Ma, Hui Li Ma

Keywords:

Burst Detection, Data Stream, Energy Ratio, TF-PDF, Topic Extraction

Export:

RIS, BibTeX

Price:

Permissions CCC:

Request Permissions

Permissions PLS:

Request Permissions

Сopyright:

Citation:

References

[1] G. Salton, C. Buckley. Term-Weighting Approaches in Automatic Text Retrieval. Information Processing and Management, 1989, 4(5): 513-523.

DOI: 10.1016/0306-4573(88)90021-0

Google Scholar

[2] Bun K. K, Ishizuka M. Topic Extraction from News Archive Using TF_PDF Algorithm. In: Proceeding of 3rd International Conference on Web Information Systems Eng. (WISE '02), 2002. 73-82.

DOI: 10.1109/wise.2002.1181645

Google Scholar

[3] Chen K Y, Luesukprasert L, Chou T. Hot Topic Extraction Based on Timeline Analysis and Multidimensional Sentence Modeling. IEEE Transactions on Knowledge and Data Engineering, 2007, 19(8): 1016-1026.

DOI: 10.1109/tkde.2007.1040

Google Scholar

[4] Qi H, Chang K Y, Lim E P. Using Burstiness to Improve Clustering of Topics in News Streams. In: Proceedings of IEEE Conference on Data Mining, 7(1): 493-498, (2007).

DOI: 10.1109/icdm.2007.17

Google Scholar

[5] Kleinberg J. Bursty and hierarchical structure in streams. Data Mining and Knowledge Discovery, 7(4): 159-178. (2003).

Google Scholar

[6] Kumar R, Novak J, Raghavan P and Tomkins A. Bursty and hierarchical structure in streams. World Wide Web, 2006, 8(2): 373-397.

Google Scholar

[7] Fung G P C, Yu X, Yu P S, Lu H J. Parameter Free Bursty Events Detection in Text Streams. In proceedings of the ACM International Conference on Very Large Data Bases, VLDB, 181–192. (2005).

Google Scholar

[8] Porter M. An Algorithm for Suffix Stripping. Program, 1980, 14(3): 211–218.

Google Scholar

[9] http: /www. nist. gov/speech/tests/tdt/, (2004).

Google Scholar