A novel proposal for outlier detection in high dimensional space

Zhana Bao, Wataru Kameyama

研究成果: Conference contribution

2 被引用数 (Scopus)

抄録

Finding rare information behind big data is important and meaningful for outlier detection. However, to find such rare information is extremely difficult when the notorious curse of dimensionality exists in high dimensional space. Most of existing methods fail to obtain good result since the Euclidean distance cannot work well in high dimensional space. In this paper, we first perform a grid division of data for each attribute, and compare the density ratio for every point in each dimension. We then project the points of the same area to other dimensions, and then we calculate the disperse extent with defined cluster density value. At last, we sum up all weight values for each point in two-step calculations. After the process, outliers are those points scoring the largest weight. The experimental results show that the proposed algorithm can achieve high precision and recall on the synthetic datasets with the dimension varying from 100 to 10000.

本文言語English
ホスト出版物のタイトルTrends and Applications in Knowledge Discovery and Data Mining - PAKDD 2013 International Workshops
ホスト出版物のサブタイトルDMApps, DANTH, QIMIE, BDM, CDA, CloudSD, Revised Selected Papers
ページ307-318
ページ数12
DOI
出版ステータスPublished - 2013
イベント17th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2013 - Gold Coast, QLD, Australia
継続期間: 2013 4月 142013 4月 17

出版物シリーズ

名前Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
7867 LNAI
ISSN(印刷版)0302-9743
ISSN(電子版)1611-3349

Conference

Conference17th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2013
国/地域Australia
CityGold Coast, QLD
Period13/4/1413/4/17

ASJC Scopus subject areas

  • 理論的コンピュータサイエンス
  • コンピュータ サイエンス(全般)

フィンガープリント

「A novel proposal for outlier detection in high dimensional space」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル