Multiple Instance Learning with Multiple Objective Genetic Programming for Web Mining

被引：15

作者：

Zafra, Amelia ^{[1
]}

Gibaja, Eva L. ^{[1
]}

Ventura, Sebastian ^{[1
]}

机构：

[1] Univ Cordoba, Dept Comp Sci & Numer Anal, Cordoba 14071, Spain

来源：

APPLIED SOFT COMPUTING | 2011年 / 11卷 / 01期

关键词：

Multi-instance learning; Multi-objective learning; Genetic programming; Web Mining; NEURAL-NETWORKS; CLASSIFICATION;

D O I：

10.1016/j.asoc.2009.10.021

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

This paper introduces a multi-objective grammar based genetic programming algorithm, MOG3P-MI, to solve a Web Mining problem from the perspective of multiple instance learning. This algorithm is evaluated and compared to other algorithms that were previously used to solve this problem. Computational experiments show that the MOG3P-MI algorithm obtains the best results, adds comprehensibility and clarity to the knowledge discovery process and overcomes the main drawbacks of previous techniques obtaining solutions which maintain a balance between conflicting measurements like sensitivity and specificity. (c) 2009 Elsevier B.V. All rights reserved.

引用

页码：93 / 102

页数：10

共 54 条

[1] Toward the next generation of recommender systems: A survey of the state-of-the-art and possible extensions
Adomavicius, G
Tuzhilin, A
[J]. IEEE TRANSACTIONS ON KNOWLEDGE AND DATA ENGINEERING, 2005, 17 (06) : 734 - 749
[2] Andrews Stuart, 2002, Proceedings of the 15th International Conference on Neural Information Processing Systems. NIPS'02, P561
[3] [Anonymous], 1997, ADV NEURAL INFORM PR
[4] [Anonymous], 2006, Nearest-Neighbor Methods in Learning and Vision: Theory and Practice Neural Information Processing
[5] [Anonymous], THESIS U TURIN TORIN
[6] [Anonymous], 2002, ICML
[7] [Anonymous], P 9 INF PROC MAN UNC
[8] [Anonymous], 2000, 17th International Conference on Machine Learning
[9] [Anonymous], 2001, INT C MACHINE LEARNI
[10] BERNSTEIN Y, 2004, GECCO 04, V3, P702

← 1 2 3 4 5 6 →