Frequent items query algorithm for uncertain sensing data

With advances in technology, large amounts of streaming data can be generated continuously by sensors. Due to the inherited limitation of sensors, these continuous sensing data can be uncertain. This calls for stream mining of uncertain sensing data. Frequent items query is valuable in wireless sensor networks (WSNs) and it can be widely used in environmental monitoring, association rules mining, and so on. A basic algorithm which continuously maintains sliding window frequent items over WSNs is proposed. However the basic algorithm needs to maintain all items in the window. Due to this, an improved algorithm is further proposed by optimizing in two aspects: (1) the pruning rules by predicting the upper bound of items probability is developed, which can reduce the candidate set and improve the query efficiency; (2) the large amount of the same items in different window can be compressed by cp-list structure in order to minimize the memory utilization. Finally, experimental results and detailed analysis demonstrate that the high efficiency and low memory cost of the proposed algorithms in WSNs.