English  |  正體中文  |  简体中文  |  Items with full text/Total items : 64194/96982 (66%)
Visitors : 8592861      Online Users : 7831
RC Version 7.0 © Powered By DSPACE, MIT. Enhanced by NTU Library & TKU Library IR team.
Scope Tips:
  • please add "double quotation mark" for query phrases to get precise results
  • please goto advance search for comprehansive author search
  • Adv. Search
    HomeLoginUploadHelpAboutAdminister Goto mobile version
    Please use this identifier to cite or link to this item: https://tkuir.lib.tku.edu.tw/dspace/handle/987654321/114704


    Title: A classification method based on interval pattern mining
    Other Titles: 基於時間間隔特徵樣式的分類方法
    Authors: 王藝筌;Wang, Yi-Chuan
    Contributors: 淡江大學資訊工程學系碩士班
    陳以錚;Chen, Yi-Cheng
    Keywords: Association Classification;classification;data mining;Interval-based Mining;sequential pattern mining;序列樣式探勘;時間間隔樣式探勘;資料探勘;關聯法則分類器
    Date: 2017
    Issue Date: 2018-08-03 15:01:19 (UTC+8)
    Abstract: 為了能夠挖掘並且瞭解大量資料中所隱含的資訊,其中是以關聯法則為基礎的方法最為著名且廣泛地運用,而最廣為人知的應用則是藉由找出大量資料間的特殊關係來進行分類以達到預測的目的。大多數研究都著重於利用時間點的資料進行序列探勘的分類法則,然而,在實際應用的例子中,不可忽略事件與事件發生時相互呼應而產生時間的關聯性、順序性,例如:電器的使用時間、患者病症發作時間。事件資料之間具有順序性的序列資料,因其大量出現於生活中,於是我們相當注重這一議題。在本篇研究當中,我們利用時間端點表示法來表示事件與事件之間的關係,採用P-TPMiner (Probabilistic Temporal Pattern Miner)來找尋所有頻繁之時間序列,整合樣式探勘與分類方法之模型,藉此制訂分類規則的計算機制,進行預測序列資料所屬之類別。從實驗結果可得知,此序列資料分類方法不只有效率且可擴性高,並且預測結果具有可靠的正確率。
    Most classification methods on sequential pattern mining are revolved about time point-based event data. Few researches utilize discovered temporal pattern for classifying. However, in many real world applications, there are relationships between events. In this paper, these relationships are simulated using a coincidence representation that extends Allen’s interval algebra. Moreover, we employ an efficient pattern mining algorithm called P-TPMiner (Probabilistic Temporal Pattern Miner) is designed to discover frequent time-interval based patterns. Exploiting the discovered temporal patterns, we proposed a classification method which is based on interval pattern. Experiments result is not only efficient and scalable, but also the accuracy is great on both synthetic and real datasets.
    Appears in Collections:[資訊工程學系暨研究所] 學位論文

    Files in This Item:

    File Description SizeFormat
    index.html0KbHTML140View/Open

    All items in 機構典藏 are protected by copyright, with all rights reserved.


    DSpace Software Copyright © 2002-2004  MIT &  Hewlett-Packard  /   Enhanced by   NTU Library & TKU Library IR teams. Copyright ©   - Feedback