Skyline Pattern Mining by Quantity-Utility Constraints in Large-Scale Databases
Jimmy Ming‐Tai Wu, Ranran Li, Jerry Chun‐Wei Lin · 2022 IEEE International Conference on Data Mining Workshops (ICDMW) · 2022
In recent years, taking into account the problem from one side has increasingly failed to meet the needs of society. Therefore, skyline quantity utility pattern mining (SQUPM) is proposed to combine two dimensions, utility and quantity, to re-veal more important information. In previous studies, researchers have proposed algorithms for small datasets, however, with the technological improvements and developments, the Internet of Things (loT), Internet of Vehicles, cell phone users, shopping websites, etc., are generating large amounts of data every day, and previous algorithms cannot handle large-scale datasets. This paper designs a three-stage MapReduce framework based on Hadoop, a big data processing platform, to mine skyline quantity utility patterns from large datasets. Experimental result shows that the algorithm is able to handle large-scale datasets and shows good performance on Hadoop clusters.