Welcome to the InfoSci Platform
Reference Hub4
Robust Classification Based on Correlations Between Attributes

Robust Classification Based on Correlations Between Attributes

Alexandros Nanopoulos, Apostolos N. Papadopoulos, Yannis Manolopoulos, Tatjana Welzer-Druzovec
Copyright: © 2007 |Volume: 3 |Issue: 3 |Pages: 14
ISSN: 1548-3924|EISSN: 1548-3932|ISSN: 1548-3924|EISBN13: 9781615202072|EISSN: 1548-3924|DOI: 10.4018/jdwm.2007070102
Cite Article Cite Article

MLA

Nanopoulos, Alexandros, et al. "Robust Classification Based on Correlations Between Attributes." IJDWM vol.3, no.3 2007: pp.14-27. http://doi.org/10.4018/jdwm.2007070102

APA

Nanopoulos, A., Papadopoulos, A. N., Manolopoulos, Y., & Welzer-Druzovec, T. (2007). Robust Classification Based on Correlations Between Attributes. International Journal of Data Warehousing and Mining (IJDWM), 3(3), 14-27. http://doi.org/10.4018/jdwm.2007070102

Chicago

Nanopoulos, Alexandros, et al. "Robust Classification Based on Correlations Between Attributes," International Journal of Data Warehousing and Mining (IJDWM) 3, no.3: 14-27. http://doi.org/10.4018/jdwm.2007070102

Export Reference

Mendeley
Favorite Full-Issue Download

Abstract

The existence of noise in the data significantly impacts the accuracy of classification. In this article, we are concerned with the development of novel classification algorithms that can efficiently handle noise. To attain this, we recognize an analogy between k nearest neighbors (kNN) classification and user-based collaborative filtering algorithms, as they both find a neighborhood of similar past data and process its contents to make a prediction about new data. The recent development of item-based collaborative filtering algorithms, which are based on similarities between items instead of transactions, addresses the sensitivity of user-based methods against noise in recommender systems. For this reason, we focus on the item-based paradigm, compared to kNN algorithms, to provide improved robustness against noise for the problem of classification. We propose two new item-based algorithms, which are experimentally evaluated with kNN. Our results show that, in terms of precision, the proposed methods outperform kNN classification by up to 15%, whereas compared to other methods, like the C4.5 system, improvement exceeds 30%.

Request Access

You do not own this content. Please login to recommend this title to your institution's librarian or purchase it from the IGI Global bookstore.