Most existing sequential pattern mining algorithms are hard to discover long meaningful sequential patterns in data stream. In this paper, the authors present BSMDS algorithms, a new approach for mining approximate sequential pattern in data stream, which is based on bitmap and multiple sequence alignment. A new bitmap-based similarity between sequences is defined to cluster similar sequences in the batch. Moreover, a bitmap-based multiple sequence alignment technology is introduced to discover long meaningful sequential patterns. In the initial stage of the algorithm, the data stream is divided into batches.