A multi-level ant-colony mining algorithm for membership functions

Hong, Tzung-Pei; Tung, Ya-Fang; Wang, Shyue-Liang; Wu, Yu-Lung; Wu, Min-Thai

doi:10.1016/j.ins.2010.12.019

Cited by 19 publications

(9 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Besides many real world transactions consist of quantitative attributes. That is why numerous researchers have been serving on the generation of association rules for quantitative data [2] [3].…”

Section: Introductionmentioning

confidence: 99%

“…In other words, modern algorithms neglect or exaggerate items beside the boundary. When distributing an attribute in the data into sets comprising individual ranges of values, the users are faced with the sharp boundary problem [2]. Quantitative attributes are discretized by joining the concept of hierarchies.…”

Section: Introductionmentioning

confidence: 99%

“…This manner occurs before the event of mining. For example, a concept hierarchy for age may be adopted to reconstruct the initial numeric values of this attribute by ranges [2]. To surmount this problem researchers are laboring for mining association rules for quantitative attributes.…”

Section: Introductionmentioning

confidence: 99%

“…Furthermore, the needs for fuzzy membership function to be given by an expert, which is not always straightforward and can be biased. Despite that, fuzzy association rule mining approach appeared out of the requirement to mine quantitative data uniformly present in databases efficiently [2]. There are two essential basic criteria for association rules, support(s) and confidence(c).…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Novel Approach for Discovery Quantitative Fuzzy Multi-Level Association Rules Mining Using Genetic Algorithm

Darwish¹,

Amer²,

Taktak³

2016

ijarai

View full text Add to dashboard Cite

Abstract-Quantitative multilevel association rules mining is a central field to realize motivating associations among data components with multiple levels abstractions. The problem of expanding procedures to handle quantitative data has been attracting the attention of many researchers. The algorithms regularly discretize the attribute fields into sharp intervals, and then implement uncomplicated algorithms established for Boolean attributes. Fuzzy association rules mining approaches are intended to defeat such shortcomings based on the fuzzy set theory. Furthermore, most of the current algorithms in the direction of this topic are based on very tiring search methods to govern the ideal support and confidence thresholds that agonize from risky computational cost in searching association rules. To accelerate quantitative multilevel association rules searching and escape the extreme computation, in this paper, we propose a new genetic-based method with significant innovation to determine threshold values for frequent item sets. In this approach, a sophisticated coding method is settled, and the qualified confidence is employed as the fitness function. With the genetic algorithm, a comprehensive search can be achieved and system automation is applied, because our model does not need the userspecified threshold of minimum support. Experiment results indicate that the recommended algorithm can powerfully generate non-redundant fuzzy multilevel association rules.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

A Novel Approach for Discovery Quantitative Fuzzy Multi-Level Association Rules Mining Using Genetic Algorithm

Darwish¹,

Amer²,

Taktak³

2016

ijarai

View full text Add to dashboard Cite

show abstract

“…As contribuições se diferenciam na construção dos grafos e nos meios de caminhar no grafo, que tendem a gerar diferentes regras. Também tem algumas outras abordagens como o trabalho de Besemann, Wu e Denton (2011), que utiliza uma variação do Apriori aplicado em redes, e o trabalho de Hong et al (2012), que utiliza um sistema de otimização de formigueiro para geração de regras. Os trabalhos propostos na extração de regras de associação visam uma modelagem otimizada com técnicas que já direcionam a geração das regras para as regras mais interessantes.…”

Section: Abordagens De Extração De Regrasunclassified

Redes de regras de associação

Padua¹

View full text Add to dashboard Cite

Association rules are widely used in the literature to extract and explore correlations within databases. The rules are extracted through a combinatorial analysis of all possible variable values, ranging in size from 2 to N, and filtered by measures such as support and confidence. Support applies a minimum occurrence filter, while confidence has a minimum conditional probability filter. For this reason, association rules tend to present 1 of 2 problems: (i) the values of support and confidence are too high and only obvious rules are presented or (ii) the values of support and confidence are too low and the number of extracted rules is extremely high. In case (i), the extracted knowledge is probably not new to the area expert, which makes the entire mining process non-productive. In case (ii), there is potentially useful knowledge extracted by the rules; However, due to the high number of standards, this knowledge is difficult to find. In order to assist the problem described in (ii), some association rule postprocessing approaches have been proposed, among them the Association Rule Network (ARN). The ARN is able to explore the rule base according to an objective item, focusing all exploration on identifying which base items correlate with the chosen item. When modeling only a single item, the ARN proved incomplete, since dominant items can relate to multiple items in a database but are not important for any of them to occur. In this doctorate we proposed 2 approaches capable of exploring the generated rules, focusing the exploration on more than one objective item. The Conventional exARN and the Greedy exARN. By exploring rules with more than one objective item, the proposed approaches are able to identify dominant items, which are items that relate to multiple objective items, and determining items, which relate to only a single objective item. The results for both approaches were promising. The Conventional exARN performed well on a less dense bases, where there are fewer relationships between items. The Greedy exARN has performed well on extremely dense bases, since the greedy algorithm behind the approach is able to drastically reduce the amount of rules modeled.

show abstract