An Efficient Method for Mining Erasable Itemsets Using Multicore Processor Platform

Huynh, Bao; Vo, Bay

doi:10.1155/2018/8487641

Cited by 7 publications

(4 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…So, we need to research efficient algorithms to solve the problem. After Deng first proposed the META [1] method to mine erasable itemsets in 2009, researchers have proposed several improved methods like VME [2], MERIT [3], MERIT+ [4], dMERIT+ [4], MEI [5], sMEI [6], EIFDD [7], MEIC [8], pMEI [9], and MSPPC [10]. These algorithms can be highly efficient in mining erasable itemsets from product databases.…”

Section: Introductionmentioning

confidence: 99%

A Parallel Mining Algorithm for Maximum Erasable Itemset Based on Multi-core Processor

2024

Teh. vjesn.

View full text Add to dashboard Cite

Mining the erasable itemset is an interesting research domain, which has been applied to solve the problem of how to efficiently use limited funds to optimise production in economic crisis. After the problem of mining the erasable itemset was posed, researchers have proposed many algorithms to solve it, among which mining the maximum erasable itemset is a significant direction for research. Since all subsets of the maximum erasable itemset are erasable itemsets, all erasable itemsets can be obtained by mining the maximum erasable itemset, which reduces both the quantity of candidate and resultant itemsets generated during the mining process. However, computing many itemset values still takes a lot of CPU time when mining huge amounts of data. And it is difficult to solve the problem quickly with sequential algorithms. Therefore, this proposed study presents a parallel algorithm for the mining of maximum erasable itemsets, called PAMMEI, based on a multi-core processor platform. The algorithm divides the entire mining task into multiple subtasks and assigns them to multiple processor cores for parallel execution, while using an efficient pruning strategy to downsize the space to be searched and increase the mining speed. To verify the efficiency of the PAMMEI algorithm, the paper compares it with most advanced algorithms. The experimental results show that PAMMEI is superior to the comparable algorithms with respect to runtime, memory usage and scalability.

show abstract

Section: Introductionmentioning

confidence: 99%

A Parallel Mining Algorithm for Maximum Erasable Itemset Based on Multi-core Processor

2024

Teh. vjesn.

View full text Add to dashboard Cite

show abstract

“…Experiments proved that the proposed approach improved the speedup and efficiency of the algorithm. Huynh and Vo [33] employed a similar technique to mine erasable itemsets. A search tree is constructed and each bough is considered to be a separate task.…”

Section: Introductionmentioning

confidence: 99%

A multithreaded hybrid framework for mining frequent itemsets

Poovan

Acharya

Reddy³

2022

IJECE

View full text Add to dashboard Cite

<p><span>Mining frequent itemsets is an area of data mining that has beguiled several researchers in recent years. Varied data structures such as Nodesets, DiffNodesets, NegNodesets, N-lists, and Diffsets are among a few that were employed to extract frequent items. However, most of these approaches fell short either in respect of run time or memory. Hybrid frameworks were formulated to repress these issues that encompass the deployment of two or more data structures to facilitate effective mining of frequent itemsets. Such an approach aims to exploit the advantages of either of the data structures while mitigating the problems of relying on either of them alone. However, limited efforts have been made to reinforce the efficiency of such frameworks. To address these issues this paper proposes a novel multithreaded hybrid framework comprising of NegNodesets and N-list structure that uses the multicore feature of today’s processors. While NegNodesets offer a concise representation of itemsets, N-lists rely on List intersection thereby speeding up the mining process. To optimize the extraction of frequent items a hash-based algorithm has been designed here to extract the resultant set of frequent items which further enhances the novelty of the framework.</span></p>

show abstract

“…In other words, we wish to find the sets of itemsets which can best be eliminated (erased) (called EPs) so that managers can then utilize this knowledge to create a new production plan that minimizes the profit reduction. Numerous algorithms have been proposed to solve the EP mining problem, such as META [12], MEI (mining erasable itemsets) [19], EIFDD (erasable itemsets for very dense datasets) [20], pMEI (parallel mining erasable itemsets) [21], and BREM (bitmap-representation erasable mining) [22]. Several variations have also been developed, such as mining EPs with constraints [23], mining erasable closed patterns [24], mining maximal EPs [25], mining top-rank-k EPs [26,27] mining erasable patterns in incremental database [28,29], and mining EPs in data streams [30][31][32][33].…”

Section: Introductionmentioning

confidence: 99%

A Fast Algorithm for Mining Top-Rank-k Erasable Closed Patterns

Nguyen¹,

Le²

2022

Computers, Materials &Amp; Continua

View full text Add to dashboard Cite

The task of mining erasable patterns (EPs) is a data mining problem that can help factory managers come up with the best product plans for the future. This problem has been studied by many scientists in recent times, and many approaches for mining EPs have been proposed. Erasable closed patterns (ECPs) are an abbreviated representation of EPs and can be considered condensed representations of EPs without information loss. Current methods of mining ECPs identify huge numbers of such patterns, whereas intelligent systems only need a small number. A ranking process therefore needs to be applied prior to use, which causes a reduction in efficiency. To overcome this limitation, this study presents a robust method for mining toprank-k ECPs in which the mining and ranking phases are combined into a single step. First, we propose a virtual-threshold-based pruning strategy to improve the mining speed. Based on this strategy and dPidset structure, we then develop a fast algorithm for mining top-rank-k ECPs, which we call TRK-ECP. Finally, we carry out experiments to compare the runtime of our TRK-ECP algorithm with two algorithms modified from dVM and TEPUS (Top-rank-k Erasable Pattern mining Using the Subsume concept), which are state-of-the-art algorithms for mining top-rank-k EPs. The results for the running time confirm that TRK-ECP outperforms the other experimental approaches in terms of mining the top-rank-k ECPs.

show abstract

An Efficient Method for Mining Erasable Itemsets Using Multicore Processor Platform

Cited by 7 publications

References 29 publications

A Parallel Mining Algorithm for Maximum Erasable Itemset Based on Multi-core Processor

A Parallel Mining Algorithm for Maximum Erasable Itemset Based on Multi-core Processor

A multithreaded hybrid framework for mining frequent itemsets

A Fast Algorithm for Mining Top-Rank-k Erasable Closed Patterns

Contact Info

Product

Resources

About