Special Issue Article
Mining Frequent Patterns with Screening of Null Transactions Using Different Models
|B.Subbulakshmi1, A. Periya Nayaki2, Dr. C. Deisy3
|Related article at Pubmed, Scholar Google|
Frequent Pattern Mining plays an important role in the field of data mining community today. The concept of frequent pattern mining can be extended to dynamic databases and data streams. A data stream represents a massive input data that arrives at high speed and is unbounded. There are various data processing models in data streams. The challenge in frequent pattern mining is the presence of null transactions. Null transaction is a transaction that does not contain any itemset being examined. Most of the existing streaming algorithms did not consider the overhead of null transactions. Hence, they fail to discover the frequent patterns faster and occupy lot of memory space to represent frequent items. To overcome this issue, a new algorithm called Screening of Null Transactions-Frequent Pattern Mining over Data Streams (SNT-FPMoDS) has been proposed which extracts frequent patterns using landmark and sliding window models. Experimental results using real datasets on different models show that our proposed algorithm saves lot of computation time and memory.