一种泛化约束的(K,L)-匿名算法

针对匿名算法在泛化过程中存在过度泛化从而导致数据可用性降低的问题,提出了一种基于泛化约束的(K,L)匿名算法.为了保证数据的可用性,该算法首先通过MAGVal(maximum alowed generalization value)形成MAM(maximum allowed microdata)来控制准标识属性值的边界范围,然后再进行(K,L)匿名的筛选,从而达到最终匿名保护的目的.实验表明,该算法在保证信息可用性的同时,减小了泛化的时间开销.     

Structure and formation process of (K,Na)-clinoptilolite

In order to synthesize (K,Na)-clinoptilolite, a (K,Na)-aluminosilicate slurry with a seed crystal was stirred at 30 °C for 4 days and then hydrothermally treated under homogeneous mixing conditions at 120 °C for 14 days. The Si/Al:(Na+K)/Si:K/(Na+K):H₂O/Al molar ratio of the starting materials was 6.0:0.42:0.5:52.5. The formation process of (K,Na)-clinoptilolite was investigated by a variety of techniques. When the materials were hydrothermally treated up to 6 days, amorphous (K,Na)-aluminosilicate was formed. After the hydrothermal treatment for 8 days, irregular particles due to the amorphous (K,Na)-aluminosilicate disappeared and plate-like particles of (K,Na)-clinoptilolite were formed, indicating that the amorphous aluminosilicate crystallized as clinoptilolite. In addition, (Na+K)/Al molar ratio in the materials varied from 1.2 to 1.0, accompanying the decrease of cation exchange capacity. During the crystallization, the wavenumber of external linkages of TO bonds shifted to higher values, implying that the SiOSi and SiOAl bonds were formed.