![]() | Only 14 pages are availabe for public view |
Abstract The increasing prominence of data streams arising in a wide range of advanced applications has led to the study of online mining of frequent itemsets. Unlike mining static databases, mining data streams poses many new challenges. The high complexity of the frequent itemsets mining problem hinders the application of the stream mining techniques. In this work, a new mining model is presented to extract frequent itemsets frorri online data stream based on a hi erarchal Partial order set (POset) representation of the generated itemsets. This model is flexible enough to deal with the whole data stream or only with the recent incoming portions. Its structure can also cope with the expiration nature of data stream by maintaining the expired POsets into the secondary memory. This flexible structure also enables expanding it as a variation of the model by using the POset structure to handle the entering transactions as the actual occurring itemsets in an online phase then generating the corresponding embedded itemsets in an offline phase. This model shows a reasonable performance after a fair comparison with other analogous successful algorithms. The main advantage of this model is the extraction of exact frequent itemsets on the whole data stream not only the recent portion. The second advantage is gained from its variation, which is the high speed achieved in the online phase of the |