A parallel algorithm for mining constrained frequent patterns using MapReduce
| Autor: | Jifu Zhang, Xiaowu Yan, Xiao Qin, Yaling Xun |
|---|---|
| Rok vydání: | 2015 |
| Předmět: |
Association rule learning
business.industry Computer science Parallel algorithm 020207 software engineering Computational intelligence 02 engineering and technology Machine learning computer.software_genre Theoretical Computer Science Tree (data structure) Scalability 0202 electrical engineering electronic engineering information engineering Programming paradigm Key (cryptography) 020201 artificial intelligence & image processing Geometry and Topology Artificial intelligence Data mining business computer Software |
| Zdroj: | Soft Computing. 21:2237-2249 |
| ISSN: | 1433-7479 1432-7643 |
| Popis: | Constrained frequent pattern refers to a frequent pattern generated using constrained conditions given by users and has characteristics of stronger pertinence, higher practicability and mining efficiency, etc. With the increasing of datasets, there are defects during the construction of the constrained frequent pattern tree, so that the constrained frequent pattern tree is difficult to apply to massive datasets. In this paper, a parallel mining algorithm of the constrained frequent pattern, called PACFP, is proposed using the MapReduce programming model. First, key steps in the algorithm, such as mapping transaction in datasets to frequent item support count, constructing the constrained frequent pattern tree, generating the constrained frequent pattern, and aggregating frequent patterns, are implemented by three pairs of Map and Reduce functions. Second, migration of data recording is achieved by applying a data grouping strategy based on frequent item support, and load balance is effectively solved while generating the constrained frequent pattern. In the end, experimental results validate availability, scalability, and expandability of the algorithm using celestial spectrum datasets. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink