QAR-CIP-NSGA-II: A new multi-objective evolutionary algorithm to mine quantitative association rules

Martín, Daniel; Rosete, Alejandro; Alcalá‐Fdez, Jesús; Herrera, F.

doi:10.1016/j.ins.2013.09.009

Cited by 66 publications

(27 citation statements)

References 50 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The fitness function maximizes the correlation between support and confidence. The authors in [21] proposed a MOEA based on the NSGA-II scheme which optimizes the product between the support and confidence, the interest of the rules and the comprehensibility. Furthermore, the amplitude of the rules was also evaluated.…”

Section: Related Workmentioning

confidence: 99%

“…Section 5.1 describes the datasets from the public BUFA repository, in which MOQAR has been tested. Then, a summary of the main parameter settings of MOQAR and the MOEA (MODENAR [5], the MOEA proposed in [15] henceforth named MOEA_Ghosh, and the MOEA described in [21] hereinafter called QAR-CIP-NSGAII) analyzed in the comparative study can be found in Sect. 5.2.…”

Section: Experimentationmentioning

confidence: 99%

“…crossover = 0.8 Difference threshold = 5 Amplitude factor = 2 W sup = 0.8 Point crossover = 2 Prob. mutation = 0.1 rule mining using genetic algorithms) and QAR-CIP-NSGAII [21] algorithms, which are available in KEEL software. It is noteworthy that the parameter settings of the studied algorithms are those appearing as default parameter values in the KEEL tool, having followed the kind suggestions of the authors.…”

Section: Parameters Setup Of Moqar and The Algorithms Used In The Commentioning

confidence: 99%

“…This section details a comparative study between the best configuration settings of MOQAR, MODENAR [5], MOEA_Ghosh [15] and QAR-CIP-NSGAII [21]. As a result of the previous section, eDominance and indDominance have been the distances used in the secondary rank- when using the eDominance and indDominance distances for the secondary ranking, respectively.…”

Section: Comparison To Other Algorithmsmentioning

confidence: 99%

“…Specifically, datasets from the public Bilkent University Function Approximation (BUFA) repository [16] have been used. The best configuration settings of objectives and distance assignments have been compared to those obtained by other existing MOEA to discover QAR, specifically, the MOEA proposed in [5,15]and [21], henceforth named MODENAR, MOEA_Gosh and QAR-CIP-NSGAII, respectively. Nonparametric statistical tests have been used in the performance study.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Improving a multi-objective evolutionary algorithm to discover quantitative association rules

et al. 2015

View full text Add to dashboard Cite

This work aims at correcting flaws existing in multi-objective evolutionary schemes to discover quantitative association rules, specifically those based on the wellknown non-dominated sorting genetic algorithm-II (NSGA-II). In particular, a methodology is proposed to find the most suitable configurations based on the set of objectives to optimize and distance measures to rank the non-dominated solutions. First, several quality measures are analyzed to select the best set of them to be optimized. Furthermore, different strate-gies are applied to replace the crowding distance used by NSGA-II to sort the solutions for each Pareto-front since such distance is not suitable for handling many-objective problems. The proposed enhancements have been integrated into the multi-objective algorithm called MOQAR. Several experiments have been carried out to assess the algorithm's performance by using different configuration settings, and the best ones have been compared to other existing algorithms. The results obtained show a remarkable performance of MOQAR in terms of quality measures.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Experimentationmentioning

confidence: 99%

Section: Parameters Setup Of Moqar and The Algorithms Used In The Commentioning

confidence: 99%

Section: Comparison To Other Algorithmsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Improving a multi-objective evolutionary algorithm to discover quantitative association rules

et al. 2015

View full text Add to dashboard Cite

show abstract

Chaos numbers based a new representation scheme for evolutionary computation: Applications in evolutionary association rule mining

Altay

Alataş

2021

Concurrency and Computation

View full text Add to dashboard Cite

In some practical situations, new computational methods are required for appropriately representing systems and their variables with inaccuracies, uncertainties, or variability. The chaos numbers seem to efficiently represent a set or range of values with lower and upper bounds for variables. In this article, a new chaos-enhanced representation scheme that is based on the notion of chaos numbers is proposed for evolutionary optimization methods. As a first application of chaos-based new encoding type, it is integrated into the novel hybrid intelligent optimization method proposed that is adapted as numerical association rules miner for the first time. The proposed hybrid method can also be used for complex types of search and optimization problems.The method is designed as a multiobjective rule miner that simultaneously handles different conflicting objectives and finds the accurate and comprehensible rules automatically. Based on the chaotic encoding, the proposed method easily and effectively adjusts the intervals of the attributes and automatically mines the rules without any preprocess. The performance of the proposed method was tested in real quantitative data sets and results were compared with the other association rule mining methods.According to the obtained results, the proposed method seems promising with respect to different metrics.

show abstract

Multi-Objective Evolutionary Optimization Algorithms for Machine Learning: A Recent Survey

Alexandropoulos

Aridas

Kotsiantis

et al. 2019

Approximation and Optimization

View full text Add to dashboard Cite

QAR-CIP-NSGA-II: A new multi-objective evolutionary algorithm to mine quantitative association rules

Cited by 66 publications

References 50 publications

Improving a multi-objective evolutionary algorithm to discover quantitative association rules

Improving a multi-objective evolutionary algorithm to discover quantitative association rules

Chaos numbers based a new representation scheme for evolutionary computation: Applications in evolutionary association rule mining

Multi-Objective Evolutionary Optimization Algorithms for Machine Learning: A Recent Survey

Contact Info

Product

Resources

About