An evolutionary algorithm to discover numeric association rules

Mata, Jacinto; Álvarez, José Luis Domínguez; Riquelme, José C.

doi:10.1145/508791.508905

Cited by 64 publications

(25 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Nevertheless, these algorithms are time intensive when applied to large datasets. For this reason, many other algorithms are based on metaheuristics like GAR [31], PSOARM [23], and BSO-ARM [11]. These methods aim to find good quality solutions for large benchmarks in a reasonable time.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

GPU-based bees swarm optimization for association rules mining

et al. 2015

View full text Add to dashboard Cite

Association rules mining (ARM) is a well-known combinatorial optimization problem aiming at extracting relevant rules from given large-scale datasets. According to the state of the art, the bio-inspired methods proved their efficiency by generating acceptable solutions in a reasonable time when dealing with small and medium size instances. Unfortunately, to cope with large instances such as the webdocs benchmark, these methods require more and more powerful processors and are time expensive. Nowadays, computing power is no longer a real issue. It can be provided by the power of emerging technologies such as graphics processing units (GPUs) that are massively multi-threaded processors. In this paper, we investigate the use of GPUs to speed up the computation. We propose two GPU-based bees swarm algorithms for association rules mining (single evaluation in GPU, SE-GPU and multiple evaluation in GPU, ME-GPU). SE-GPU aims at evaluating one rule at a time where each thread is associated with one transaction, whereas ME-GPU evaluates multiple rules in parallel on GPU where each thread is associated with several transactions. To validate our 123 Y. Djenouri et al. approaches, the two algorithms have been executed to solve well-known large ARM instances. Real experiments have been carried out on an Intel Xeon 64 bit quad-core processor E5520 coupled to an Nvidia Tesla C2075 GPU device. The results show that our approaches improve the execution time up to 100× over the sequential mono-core bees swarm optimization-ARM algorithm. Moreover, the proposed approaches have been compared with CPU multi-core ones (1-8 cores). The results show that they are faster than the multi-core versions whatever is the number of used cores.

show abstract

Section: Related Workmentioning

confidence: 99%

“…To deal with a huge transactional database in a reasonable time, many metaheuristics have already been applied to ARM. Some of these methods are based on evolutionary algorithms [31,42,48], and other ones on swarm intelligence [11,20,23,34].…”

Section: Introductionmentioning

confidence: 99%

GPU-based bees swarm optimization for association rules mining

et al. 2015

View full text Add to dashboard Cite

show abstract

“…Genetic algorithm (GA) is the first evolutionary algorithm used to solve the ARM problem, such as GENAR [18], GAR [19]. Those two methods used the standard version of the genetic algorithm with poor representation of the solutions.…”

Section: Arm With Metaheuristicsmentioning

confidence: 99%

Penguins Search Optimisation Algorithm for Association Rules Mining

Gheraibia¹

2016

CIT. J. comput. inf. technol.

View full text Add to dashboard Cite

Association Rules Mining (ARM) is one of the most popular and well-known approaches for the decisionmaking process. All classic exhaustive ARM algorithms are time consuming and generate a very large number of association rules, even the recent proposed meta-heuristics based methods generate a small number of high quality rules but with high overlapping. To deal with this issue, we propose a new ARM approach based on penguins search optimisation algorithm (Pe-ARM for short). Moreover, an efficient measure is incorporated into the main process to evaluate the amount of overlapping among the generated rules. The proposed approach also ensures a good diversification over the whole solutions space. To demonstrate the effectiveness of the proposed approach, several experiments have been carried out on different data sets and specifically on the biological ones. The results reveal that the proposed approach outperforms the well known meta-heuristics ARM algorithms in both execution time and solution quality.

show abstract

“…Some approaches do not mine all general itemsets but rather focus on finding the best intervals for a given rule [7,9,12,17,18,23,22,24]. The input to these algorithms is a rule that have some uninstantiated attributes.…”

Section: Background and Related Workmentioning

confidence: 99%

“…Another novel approach is proposed to use genetic algorithms to mine optimized association rules. Mata et al [17,18] use a genetic algorithm to optimize the support of an interval for a quantitative attribute. However, their approach is limited to datasets without any overlap among different classes of the data.…”

Section: Background and Related Workmentioning

confidence: 99%