本研究將修改 PrefixSpan 序列資料探勘方法並結合貝氏認知網路（Bayesian Belief Network）建構新的序列資料探勘方法。透過本方法以達成有效減少探勘時產生的大量映射資料庫的問題；並且可以清楚地表達所有資料中每一個項目彼此之間的先後順序和動態機率。透過貝氏認知網路除了可以清楚地將項目間的關聯以圖形的方式來呈現之外，也能能將相同領域內的知識和資料做結合利用條件機率的設定與事件的推論來加以表達。為了建構貝氏認知網路中項目間的關聯與條件機率的設定，本研究將嘗試修改PrefixSpan 序列資料探勘方法來取得相關的資訊。透過此一新的方法將可提供更豐富且具推論能力之序列資料探勘結果。使用者可設定不同項目發生之狀況並推論得知在未來可能發生或過去曾經發生之項目的機率值，以做為各種決策之用。 The probability of each event in a sequential pattern can provide more information for decision maker to analyze and predict the behavior of correlated pattern. However, in the previous studies there is no technique developed to simultaneously discover the probability in the pattern mining process. Thus, to provide such information, we will extend the PrefixSpan method and use the first and second level mining results to estimate the conditional probability of a sequence pattern event. Then, a directed graph will be constructed to build a Bayesian Belief Network (BBN) based on the second level discovered pattern and the estimated conditional probability. The created model can be used to estimate the probability of occurrence for future events if some events are observed in advance.
