持续时态数据挖掘的研究

Research on continuous temporal data mining

基于一阶线性时态逻辑,形式化定义时态数据挖掘中的主要概念,利用线性状态结构对每个时间点上的一阶语言符号进行赋值,并度量公式的真值范围.按照挖掘段概念,开发持续挖掘过程模型,用于归纳局部一阶规则与推导高阶规则.基于信息扩散原理,提出一阶规则的度量值估计方法和规则泛化算法.最后通过算例说明了扩散估计和算法的有效性.

The definitions of main notions used in temporal knowledge discovering are proposed in a formal way, which is based on first-order linear temporal logic. The concept of linear state structure allows associating each time moment with an valuation of all symbols of a first-order language, and measures the extent of truth of a formula. According to the notion of session mining, a continuous data mining process model is developed for inducing the local first-order rules and inferring higher order rules. Based on the principle of the information diffusion, the estimation for the measures and an algorithm for rule generalization are presented. The simulations show the effectiveness of the methods.