Efficient algorithms for frequent pattern mining in many-task computing environments

摘要: The goal of data mining is to discover hidden useful information in large databases. Mining frequent patterns from transaction databases an important problem mining. As the database size increases, computation time and required memory also increase. Because number items user behaviours become more complex. To solve increasing complexity, many researchers have applied parallel distributed computing techniques discovery amounts data. However, most studies focused on improving performance for a single task neglected many-task issue, which current cloud-computing environments. In these environments, application often provided as service, e.g., Google search engine, implying that users can use it simultaneously. this paper, we propose set algorithms, containing Equal Working Set (EWS) algorithm, Request On Demand (ROD) Small Size (SSWS) algorithm Progressive (PSWS) pattern provides fast scalable service Through empirical evaluations various simulation conditions, proposed algorithms are shown deliver excellent with respect scalability execution time.