Stochastic population-based nature-inspired metaheuristics have been proven as a robust tool for mining association rules. These algorithms are very scalable, as well as very fast compared with some deterministic ones that search for solutions exhaustively. Typically, algorithms for association rule mining identify a lot of rules depending, on the transaction database and number of attributes. Therefore, evaluating these rules is very complex. On the other hand, establishing the relationships between discovered association rules can be considered as a very hard problem that cannot easily be solved manually. In this paper, we propose a new algorithm based on stochastic population-based nature-inspired metaheuristics for discovering dependencies among association rules.
Paper
Full text
Discovering dependencies among mined association rules with population-based metaheuristics
Semantic Scholar · Computer Science · 2019
Abstract
Stochastic population-based nature-inspired metaheuristics have been proven as a robust tool for mining association rules. These algorithms are very scalable, as well as very fast compared with some deterministic ones that search for solutions exhaustively. Typically, algorithms for association rule mining identify a lot of rules depending, on the transaction database and number of attributes. Therefore, evaluating these rules is very complex. On the other hand, establishing the relationships between discovered association rules can be considered as a very hard problem that cannot easily be solved manually. In this paper, we propose a new algorithm based on stochastic population-based nature-inspired metaheuristics for discovering dependencies among association rules.