With the rapid development of data warehouse and OLAP techniques, the researchers begin to pay atten-tion to the data mining in the data cube. Recently, Dr. T. Imielinski etc. firstly presented the problem of the cube...With the rapid development of data warehouse and OLAP techniques, the researchers begin to pay atten-tion to the data mining in the data cube. Recently, Dr. T. Imielinski etc. firstly presented the problem of the cubegradient mining that is a generalization of association rule in data cube. In this paper, we firstly introduce the relatedconcepts of data cube and condensed cube with an emphasis. Then we introduce some interesting problems related tothe cube gradient mining including: constrained cube gradient mining and the query language of cube gradient. Final-ly, we introduce several issues on the combination of cube gradient and the condensed cube, that is, the cube gradientmining in the materialized data cube and the integration of cube gradient mining and cube browse.展开更多
近年来,隐私保护事务数据发布得到了研究者的广泛关注.事务数据的稀疏性导致个体隐私保护与数据效用性之间很难达到平衡.目前已有的方法大多是基于分组的匿名模型,但该类模型依赖于攻击者背景知识,且发布的数据无法满足事务数据分析任...近年来,隐私保护事务数据发布得到了研究者的广泛关注.事务数据的稀疏性导致个体隐私保护与数据效用性之间很难达到平衡.目前已有的方法大多是基于分组的匿名模型,但该类模型依赖于攻击者背景知识,且发布的数据无法满足事务数据分析任务的需要.针对事务数据隐私保护发布的数据安全性与效用性不足,基于差分隐私与压缩感知理论,提出一种有效的面向应用的事务数据发布策略(transaction data publish strategy,TDPS).首先构建事务数据库的完整Trie项集树,然后基于压缩感知技术对项集树添加满足差分隐私约束的噪音得到含噪Trie项集树,最后在含噪树上进行频繁项集挖掘任务.实验结果表明,TDPS不仅能很好地保护隐私,而且能有效保持数据效用性,满足事务数据分析任务对数据质量的要求.展开更多
文摘With the rapid development of data warehouse and OLAP techniques, the researchers begin to pay atten-tion to the data mining in the data cube. Recently, Dr. T. Imielinski etc. firstly presented the problem of the cubegradient mining that is a generalization of association rule in data cube. In this paper, we firstly introduce the relatedconcepts of data cube and condensed cube with an emphasis. Then we introduce some interesting problems related tothe cube gradient mining including: constrained cube gradient mining and the query language of cube gradient. Final-ly, we introduce several issues on the combination of cube gradient and the condensed cube, that is, the cube gradientmining in the materialized data cube and the integration of cube gradient mining and cube browse.
文摘近年来,隐私保护事务数据发布得到了研究者的广泛关注.事务数据的稀疏性导致个体隐私保护与数据效用性之间很难达到平衡.目前已有的方法大多是基于分组的匿名模型,但该类模型依赖于攻击者背景知识,且发布的数据无法满足事务数据分析任务的需要.针对事务数据隐私保护发布的数据安全性与效用性不足,基于差分隐私与压缩感知理论,提出一种有效的面向应用的事务数据发布策略(transaction data publish strategy,TDPS).首先构建事务数据库的完整Trie项集树,然后基于压缩感知技术对项集树添加满足差分隐私约束的噪音得到含噪Trie项集树,最后在含噪树上进行频繁项集挖掘任务.实验结果表明,TDPS不仅能很好地保护隐私,而且能有效保持数据效用性,满足事务数据分析任务对数据质量的要求.
