關聯規則是一項用用以分析資料間關聯性的研究。現今已經有許多關於名詞性資料(nominal data)的關聯規則研究，然而，卻沒有關於處理序數資料(ordinal data)的研究。除此之外，先前的研究通常假設資料是精確的而且確定的，然而，這項假設條件是不符合真實世界的狀況。由於人們的疏失、儀器測量或是紀錄上的限制導致無法精準紀錄精準的資料，所以真實世界的資料是不精確或是不確定的。因此，本研究提出一個結合資料探勘、模糊集合(Fuzzy sets)、可能性理論(Possibility theory)等技術的探勘方式，從不確定性資料中挖掘關聯規則(Discovering Association Rules from Imprecise or Uncertain Data, DARIUD)，用以引導如何從不精確或是不確定的的資料中挖掘出有趣、多資料型態、高度確定性的關聯規則。本研究提出三種資料型態的研究，來證明這個新探勘方式的可行性(Workable)和其一般化(Generalization)，並引導這些領域結合的新研究。 Association rule mining is an emerging data analysis method that can discover associations within data. Although there have been numerous studies on finding association rules from nominal data, few have tried to do so from ordinal data. Additionally, previous mining algorithms usually assume that the input data is precise and certain. Unfortunately, real-world data tends to be uncertain due to human errors, instrument errors, recording errors, and so on. Therefore, a question arises immediately is how we can mine association rules from imprecise or uncertain data. Therefore, this study devotes to proposing a work process, Discovering Association Rules from Imprecise or Uncertain Data (DARIUD), to hold more general viewpoint combining Data Mining, Fuzzy Sets and Possibility theory fields for discovering interesting and certain patterns. The purpose of the process is to establish a cooperative relationship to understand and analyze the investigating steps of association rules mining from imprecise or uncertain Data. Three researches were proposed to demonstrate that the process, DARIUD, can be workable and capable of generalization to future studies in the three fields.
