發展完整關聯法則於生物資訊系統

傳送到

題名:
發展完整關聯法則於生物資訊系統
著者: 鄭吉宏; Jheng, Ji-hong
孫光天; 數位學習科技學系碩士班
主題: 生物資訊; 關連法則; Apriori; 邏輯關聯; 資料探勘; Bioinformatics; Association rule; Apriori; Data mining
描述: 資料探勘技術在生物與醫學資訊系統中常被用來探勘隱藏的訊息或作自動化決策處理等相關應用。在資料探勘中，關連法則探勘是一種從大量的資料理面，探勘每一個資料項目之間是否具有特性規律的技術。關連法則在生物資訊中常用於生物序列分析等，可以尋找出不同核苷酸或是胺基酸之間是否具有關聯性。但關連法則演算法的設計，並非是針對生物序列資料的特性設計。除此之外，在有些狀況下會因為最小支持度的關係而刪除含有不頻繁項目的序列，但該序列可能隱含重要生物意義與訊息。有些情況則又面臨太多規則產生，導致生物實驗者無法直接根據探勘結果做生物實驗。因此必須有另一種關連法則演算法，具有綜整全部規則的特性。本研究將生物序列資料導入布林代數及相關運算，以簡化規則之間邏輯關係的方式，發展適用於探勘生物序列資料的完整關聯法則演算法，得以發掘生物序列隱藏的訊息。
Data mining technology has often used in finding hidden information and automatic decision of biological and medical information systems. In data mining, association rule is a technology of mining the items if including special regular patterns in large database. Association rule applied in bioinformatics was analyzed biological sequences to search some relation between different nucleotide or protein sequences. But the design of association rule was not on biological sequence data properties. In some case, the algorithm would delete the sequences that including important biological message according to minimal support value. In other case, it would produce so many rules that research workers could not experiment on biological sequences. We must develop other mining association rule algorithm to Integrating whole rule qualities. We use Boolean algebra operations to deal with biological sequences. It reduces the logical relation of each rules and developes complete association rules algorithm for mining biological sequences that including hidden messae.
碩士
建立日期: 2008
格式: 121 bytes
text/html
語言: 中文
識別號: http://nutnr.lib.nutn.edu.tw/handle/987654321/3407
資源來源: NUTN IR

連結

連回原始介面

發展完整關聯法則於生物資訊系統

鄭吉宏; Jheng, Ji-hong 孫光天; 數位學習科技學系碩士班 2008

正在檢索遠程資料庫，請稍等