| Summary: | 碩士 === 中興大學 === 資訊科學與工程學系所 === 99 === The technique of classification can classify data into different categories. With the development of information technology, the demand for sequence data classification increases. Many interesting applications involve decision prediction based on the sequence data. A sequence is an ordered list of elements. The traditional classification methods are not suitable for sequence data. Therefore, this thesis proposed a sequence data classifier model based on the sequential patterns’ length. In addition to integrating sequential pattern mining and classification techniques, this study also proposed a classification rule selection mechanism, that predicts the class label of sequence data based on pattern scores. From the experimental results, the proposed sequence data classifier model shows good performance on the synthetic and real sequence data.
|
|---|
