作者: Mansurul Bhuiyan , Mohammad Al Hasan
DOI:
关键词: Software 、 Theoretical computer science 、 Computation 、 Exploratory data analysis 、 Source code 、 Computer science 、 Graph (abstract data type) 、 Data mining 、 Data mining algorithm 、 Graph 、 Data structure
摘要: Frequent subgraph mining (FSM) is an important task for exploratory data analysis on graph data. Over the years, many algorithms have been proposed to solve this task. These assume that structure of small enough fit in main memory a computer. However, as real-world grows, both size and quantity, such assumption does not hold any longer. To overcome this, some database-centric methods recent years solving FSM; however, distributed solution using MapReduce paradigm has explored extensively. Since, becoming de- facto computation massive data, efficient FSM algorithm huge demand. In work, we propose frequent called MIRAGE which uses iterative based framework. complete it returns all subgraphs given user-defined support, applies optimizations latest adopt. Our experiments with real life large synthetic datasets validate effectiveness from datasets. The source code available www.cs.iupui.edu/alhasan/software/