| Publisher | Association for Computing Machinery | ||
|---|---|---|---|
| Format | 228.1KB PDF | Date added | 11 Nov 2005 |
| Topics | Artificial Intelligence, Streaming Media, Software Engineering | ||
| Downloads | 67 | ||
Concept drifting is an important and challenging research issue in the field of machine learning. This paper mainly addresses the issue of semantic concept drifting in time series such as video streams over a relatively long period of time. An Online-Optimized Incremental Learning framework is proposed as an example learning system for tracking the drifting concepts. Furthermore, a set of measures are defined to track the process of concept drifting in the learning system. These tracking measures are also applied to determine the corresponding parameters used for model updating in order to obtain the optimal up-to-date classifiers.
Did you find this white paper useful?
30 out of 50 users found this white paper useful
Related white papers
Autonomous Mental Development: A New Interdisciplinary Transactions for Natural and Artificial Intelligence
Although some baby animals can get up and walk within hours after birth, what a human child learns during the first two years of life easily exceeds what those animals...
Creating and Using Oracle VM Templates: The Fastest Way to Deploy Any Enterprise Software
The intended audience for this paper is end-users of Oracle VM as well as third party ISVs (Independent Software Vendors), system-integrators, and value-added resellers interested in creating and using Oracle...
Detecting a Denial of Service Using Artificial Intelligent Tools, Genetic Algorithm
This paper describes novel work in using Genetic Algorithm for detecting misuse of programs. A brief overview of Intrusion Detection System, genetic algorithm and related detection techniques is presented. Developing...
Rule-Based Anomaly Detection on IP Flows
Rule-based packet classification is a powerful method for identifying traffic anomalies, with network security as a key application area. While popular systems like Snort are used in many network locations,...
MSDN Webcast: Game Programming in DarkGDK and C++ (Part 6 of 6): Finalizing Levels, Testing, and Polish (Level 100)
The attendee of this webcast will learn how to set the difficulty curve of the game and make levels. The presenter also discusses testing the game, making final adjustments to...
MSDN Webcast: Game Programming in DarkGDK and C++ (Part 5 of 6): Explosions, Effects, and Information Displays (Level 100)
The presenter of this webcast covers the explosion effects in the game and examines how sounds are triggered by the various game elements. The presenter also explores how to display...
MSDN Webcast: Introducing Xbox Community Games (Part 1 of 2) (Level 200)
The attendee of this webcast will learn about Xbox LIVE Community Games and how to join up with the XNA Creators Club to create and sell one's own video game...
