Outlier mining on multiple time series data in stock market

Publication Type:
Conference Proceeding
Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 2008, 5351 LNAI pp. 1010 - 1015
Issue Date:
Filename Description Size
Thumbnail2008001378OK.pdf1.28 MB
Adobe PDF
Full metadata record
With the dramatic increase of stock market data, traditional outlier mining technologies have shown their limitations in efficiency and precision. In this paper, an outlier mining model on stock market data is proposed, which aims to detect the anomalies from multiple complex stock market data. This model is able to improve the precision of outlier mining on individual time series. The experiments on real-world stock market data show that the proposed outlier mining model is effective and outperforms traditional technologies. © 2008 Springer Berlin Heidelberg.
Please use this identifier to cite or link to this item: