一种面向不确定数据流的模体发现算法

借鉴生物信息学中序列模式发现思想，提出了基于MEME（multiple expectation-maximization for motif elicitation）的不确定数据流模体发现算法。该算法根据不确定数据流的特点，设计了不确定滑动窗口的简化计算方法，改进了SAX（symbolic aggregate approximation）的符号化策略，用防空反导情报传感器网络中的一组不确定数据流验证了其可行性，通过植入不同数目模体的方法测试了其准确性，并在元组存在概率为1的条件下与已有算法进行比较，验证其有效性。

New Motif Discovery Algorithm for Uncertain Data Stream

A new MEME-based motif discovery algorithm for uncertain data stream is proposed by using the idea of sequential pattern discovery in bioinformatics. According to features of uncertain data stream, the new algorithm designs a simplified calculation method for uncertain sliding window and modifies the SAX symbolic strategy. The feasibility of the proposed algorithm is verified by one uncertain test data stream from air and missile defense sensors. And its accuracy is measured through planting different number motifs. Furthermore, the proposed algorithm is validated by comparing with existing algorithms in the condition that the existence probability of tuples is set to 1.