期刊名称：International Journal of Multimedia and Ubiquitous Engineering
印刷版ISSN：1975-0080
出版年度：2020
卷号：15
期号：1
DOI：10.21742/ijmue.2020.15.1.03
语种：English
出版社：SERSC
摘要：Frequent subgraph extraction from a substantial number of small graphs is a crude activity for some, information mining applications. To extricate frequent subgraphs, existing systems need to identify countless which is super straight with the cardinality of the dataset. Given the huge developing volume of graph information, it is hard to play out the regular subgraph extraction on a unified machine proficiently. Along these lines, there is a need to explore how to effectively play out this extraction over expansive datasets utilizing MapReduce. Parallelizing existing strategies straightforwardly utilizing MapReduce does not yield great execution as it is hard to adjust the remaining task at hand among the figure hubs. This structure receives the MRFSE procedure to iteratively remove Frequent subgraphs, i.e., all incessant size-(i+1) subgraphs are created dependent on continuous size-I subgraphs at the ith emphasis utilizing a solitary MapReduce work. To productively separate successive subgraphs, arrangement and mining stage are utilized which incorporates isomorphism testing to wipe out copy designs. Frequent subgraphs extraction should be possible productively and effectively by utilizing a disseminated domain named Hadoop MapReduce structure.
