An Incremental Class Boundary Preserving Hypersphere Classifier
Authors
Abstract
Recent progress in sensing, networking and data management has led to a wealth of valuable information. The challenge is to extract meaningful knowledge from such data produced at an astonishing rate. Unlike batch learning algorithms designed under the assumptions that data is static and its volume is small (and manageable), incremental algorithms can rapidly update their models to incorporate new information (on a sample-by-sample basis). In this paper we propose a new incremental instance-based learning algorithm which presents good properties in terms of multi-class support, complexity, scalability and interpretability. The Incremental Hypersphere Classier (IHC) is tested in well-known benchmarks yielding good classication performance results. Additionally, it can be used as an instance selection method since it preserves class boundary samples.
Keywords
Incremental Learning, Classification, Machine Learning
Subject
Incremental learning
Related Project
BIOINK- Incremental Kernel Learning for Biological Data Analysis
Conference
B.-L. Lu, L. Zhang, and J. Kwok (Eds.):ICONIP 2011, Part II, LNCS 7063, pp. 690--699. Springer, Heidelberg (2011), November 2011
DOI