On the design of hardware-software architectures for frequent itemsets mining on data streams

A Review of Supervised Classification based on Contrast Patterns: Applications, Trends, and Challenges

“…Later, in 2018, the authors of [24] proposed to extend the SysTreeMiner algorithm. This extension includes a new algorithm for the sliding window model and pre-processing stage.…”

“…Also, the authors proposed new parallel algorithms for mining patterns on data streams. None of the proposals of [24] compared their results against the algorithm (EPDSminer) proposed by [7] in the same context.…”

“…The authors showed how D2P-Apriori improves other three state-of-the-art serial implementations. Nevertheless, they did not compare their results against the algorithms proposed by [24] in the same context.…”

“…The authors showed how Floating Top-k improves other well-known algorithms for mining patterns from data streams. Nevertheless, Floating Top-k was not compared against the algorithms proposed by [23][24][25] in this context.…”

“…In the same vein, many algorithms for mining CPs from data streams have been proposed [23][24][25][26]83]. Nevertheless, they continue using an initial discretization and an exhaustive-search-based algorithm for mining CPs, which are limiting the power of these proposals.…”