基于两个矩阵的关联规则挖掘优化算法

针对传统数据挖掘算法的不足，提出基于两个矩阵的优化关联规则挖掘算法。该算法对事务数据库进行一次扫描，将其转换成两个用于存放逻辑数据的矩阵，并保留项目间的关联信息。对两个矩阵进行挖掘，基于矩阵MA得到频繁1-项集和频繁2-项集，基于矩阵MB得到最大频繁项集，其他频繁k-项集基于两个矩阵和已得频繁集获取。该算法极大减少了候选频繁集数量，挖掘过程采用逻辑运算。实验结果证明了其可行性和高效性。

Optimized Algorithm for Mining Association Rule Based on Two Matrixes

Open domain Question Answering(QA) represents a challenge of natural language processing, aiming at returning exact answers in response to natural language questions. Pattern matching is an effective strategy for QA. This paper demonstrates the technique of pattern generalization and its applications in QA. There are two important applications of pattern in QA. Pattern-based query expansion retrieves more exact snippets including the correct answer and the performance is improved. Another application is the answer extraction which applies the pattern matching approach. Experimental result on TREC data indicates that the pattern matching method is effective.