Mining Frequent and Similar Patterns with Attribute Oriented Induction High Level Emerging Pattern (AOI-HEP) Data Mining Technique

Attribute Oriented Induction High level Emerging Pattern (AOI-HEP) is a novel idea which is influenced by Attribute Oriented Induction (AOI) and Emerging Pattern (EP). AOI-HEP discovers patterns such as Total Subsumption HEP (TSHEP), Subsumption Overlapping HEP (SOHEP) and Total Overlapping HEP (TOHEP), include frequent and similar patterns. Mining TSHEP, SOHEP, TOHEP, frequent and similar patterns for each dataset is influenced by learning on high level concept in one of chosen attribute. The experiments used four datasets from UCI machine learning repository and most datasets have SOHEP but not TSHEP and TOHEP and the most rarely found were TOHEP. There are total twenty two High level Emerging Pattern (HEP) where four HEP are TSHEP, sixteen HEP are SOHEP and two HEP are TOHEP, and there are five frequent and four similar patterns from the experiments. Moreover, the experiment showed that adult and breast cancer datasets are interested to mine frequent pattern while breast cancer and IPUMS datasets are interested to mine similar pattern. However, census dataset is not interested to be mined for both frequent and similar patterns. AOI-HEP is suitable for dealing with large dataset since can handle million tuples in dataset in one digit seconds.