指出了Apriroi算法的不足,并提出改進算法:在剪枝侯選項目集的時候,采用生成頻繁項
目集和刪除具有非頻繁子集候選同時進行的方法來減少多余子項目集的生成。這樣,既提高了掃描數(shù)據(jù)庫過程中的信息獲取率,又及時刪除超集不是頻繁項目集的項集,進一步縮減項集的潛在規(guī)模,提高了頻繁項目集即關(guān)聯(lián)規(guī)則生成的效率。討論了如何利用改進的Apriori算法對評教系統(tǒng)數(shù)據(jù)進行關(guān)聯(lián)規(guī)則挖掘。
關(guān)鍵詞:Apriori 算法;評教系統(tǒng);關(guān)聯(lián)規(guī)則
Abstract: This paper analyses the lack of Apriori algorithm and presents an improved algorithm. This algorithm adopts a new method to reduce the redundant generation of sub-itemsets during pruning the candidate itemsets, which can form directly the set of frequent itemsets and eliminate candidates having a subset that is not frequent in the meantime. The result shows that it can raise the probability of obtaining information in scanning database and reduce the potential scale of itemsets. It is discussed how to apply an improved Apriori algorithm to exploring association rules of the data in Teaching Evaluation System.
Key words: apriori algorithm; teaching evaluation system; association rule