During the experiment, we reduce the attributes by using the new attribute reduction algorithm, and we then reduce the value and extract the rules on the new decision list, and the last step is classification prediction and its accurate achieves 92.6 %. 在实验的过程中,首先利用新的属性约简算法进行属性约简,然后对新的决策表(DL)进行值约简并提取规则,最后进行分类预测,预测准确率高达92.6%。
Analysis of Logical Expression of Decision List(DL) 决策表(DL)的逻辑表达式分析
Kinds of data structures, which are called Editing Decision List(DL) ( EDL ), are designed and realized to satisfy the stability of software. 为满足软件的稳定性要求,本文设计和实现了各种数据结构,统称为编辑决策列表。
The excitation controller worked at on-line identification model when the system structures have been changed and updated fuzzy decision list when the identification have been succeeded, so the system have the ability of auto-learning. 当系统进入结构改变等未知状态时,转入在线参数辨识模式,辨识成功后,更新模糊决策表(DL),使系统具有自学习功能。
The rules including those got by the decision list algorithm and those devised by humans can be applied in some applications, such as Chinese word segmentation and POS tagging. 本研究所得的消歧规则(包括由决策表(DL)算法自动获取的规则和由人工总结的规则)可为分词、标注词性等系统所应用。