.png){kind=icon}
返回
Evolutionary Fuzzy Rule-Based Methods for Monotonic Classification
.png){kind=icon}
PRE
AI
DOI：10.1109/TFUZZ.2017.2718491.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
In data science applications, it is very often to require predictive models satisfying monotonicity with respect to the explanatory variables involved in the dataset. In ordinal classification or regression, this occurs when the output variable or class label do not decrease when input variables increase, or vice versa. This problem is commonly known as monotonic classification, and most existing classification techniques are not able to manage this kind of constraints or they require first to monotonize the data. In the literature, the monotonicity has been considered in linguistic fuzzy models, fuzzy-inference methods, and fuzzy rule-based control systems. However, to the best of our knowledge, there is no fuzzy rule-based system designed to produce monotonic fuzzy rule-based models for classification problems. In this paper, we propose to incorporate some mechanisms based on monotonicity indexes for addressing such problems in two popular and competitive evolutionary fuzzy systems algorithms for classification and regression tasks: FARC-HD and FSMOGFS(e)+ TUNe. In addition, the proposals are able to handle any kind of classification dataset without the necessity of preprocessing. The quality of our approaches is analyzed using statistical analysis and comparing with well-known monotonic classifiers.
Keyword:
Data science
evolutionary algorithms
fuzzy rule-based classification systems (FRBCSs)
monotonicity constraints
ordinal classification
ordinal regression
AI总结
对已上传原文的论文进行重点信息的提取，主要内容包括：简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。
期刊
IF:
11.9
论文数:
5.0K
被引数:
2.9W
机构
引用论文
A study of statistical techniques and performance measures for genetics-based machine learning: accuracy and interpretability基于遗传学的机器学习的统计技术和性能度量的研究: 准确性和可解释性.png)
SOFT COMPUTING
SOFT COMPUTINGIF2.5
PREAI
分享收藏
A Fuzzy Association Rule-Based Classification Model for High-Dimensional Problems With Genetic Rule Selection and Lateral Tuning基于遗传规则选择和横向调整的高维问题模糊关联规则分类模型
OAAI
OAAI分享收藏
查看更多内容