SWOD

Outlier Detection and Classification

A short description  
The distribution package  
Related publications 
Authors & Aknowledgement 
Contact 

A short description

Anomaly detection and change analysis are challenging tasks in stream data mining. We have developed a method that addresses both these tasks in geophysical applications. The method, called SWOD (Sliding Window Outlier Detection), is designed for numeric data routinely sampled through a sensor network. It extends the traditional time series forecasting theory by accounting for the spatial information of geophysical data. In particular, a forecasting model is computed incrementally by accounting for the temporal correlation of data which exhibit a spatial correlation in the recent past. For each sensor the observed value is compared to its spatial-aware forecast, in order to identify the outliers. Finally, the spatial correlation of outliers is analyzed, in order to classify changes and reduce the number of false anomalies. 

The distribution package

SWOD is implemented in a Java system. It iterfaces MySQL database to read the network structure (nodes and arcs). 

jar Description
SWOD This rar bundle contains (1) swod.jar that allows us to :(i) perform incrementally the discovery of trend clusters over sliding windows, as well as (ii) detect and classify outliers in geophysical data streams (2) setup files and (3) a benchmark data steam (Intel Berkeley Temeprature)


Warning: SWOD is free for evaluation, research and teaching purposes, but not for commercial purposes. 
Please Acknowledge

Top of this page

 

Related publications

Annalisa Appice, Pietro Guccione, Donato Malerba, Anna Ciampi
Dealing with temporal and spatial correlations to classify outliers in geophysical data streams. Information Sciences 285: 162-180 

Top of this page


Project team

Contact

Name Email address Tel. number Fax
Annalisa Appice annalisa.appice@uniba.it +39 080 5443262 +39 080 5443262
 

Top of this page