Paper Info
Title
Classification of Customer Call Data in the Presence of Concept Drift and Noise
Abstract
Many of today's real world domains require online classification tasks in very demanding situations. This work presents the results of applying the CD3 algorithm to telecommunications call data. CD3 enables the detection of concept drift in the presence of noise within real time data. The application detects the drift using a TSAR methodology and applies a purging mechanism as a corrective action. The main focus of this work is to identify from customer files and call records if the profile of customers registering for a 'friends and family' service is changing over a period of time. We will begin with a review of the CD3 application and the presentation of the data. This will conclude with experimental results.
Year
DOI
Venue
2002
10.1007/3-540-46019-5_6
Software
Keywords
Field
DocType
customer file,cd3 algorithm,concept drift,real world domain,cd3 application,corrective action,call record,tsar methodology,customer call data,real time data
Real-time data,Computer science,Real-time operating system,Concept drift,Real-time computing,Timestamp,Artificial intelligence,Knowledge base,Distributed computing
Conference
Volume
ISSN
ISBN
2311
0302-9743
3-540-43481-X
Citations 
PageRank 
References 
12
0.68
15
Authors
2
Name
Order
Citations
PageRank
Michaela Black122721.53
Ray Hickey221619.24