Mining minimal constrained flow cycles from complex transaction data

摘要: Transaction data in domains such as trading records from online financial or other markets, logistics delivery registers, and many others, are being accumulated at an increasing rate. In this type of each transaction has a complex format, usually associated with attributes time, numerical quantity, parties involved, so on. Performing mining on trace record transactions may enable the extraction knowledge about implicit relationships which will benefit community different ways, for example by improving market efficiency oversight, detecting scheduling bottlenecks. However, size sets is enormous, therefore order to perform searching techniques considerations often more important than correctness. paper we develop framework embed methods speed up search algorithms goal cycles that fit given constraint predicate amount quantities can flow direction. The method shown improve significantly naive approach, suggests number directions further work.