三支决策作为一种数据处理的有效模型,自提出以来,被广泛扩展应用到粒计算、概念格及模糊数学理论。特别是与粒计算的结合是处理大数据问题的一种实用方法。本文通过三角模糊数建立一个新的复合函数,并通过这个复合函数转化形式背景中出现的文本型数据。然后,利用三支决策的思想以及模糊数学中的阈值,可以将转化后的区间有效地放大或者缩小。最后,经过反复的划分,得到一种适合实际情况的区间划分。根据本文提出的文本型数据的划分模型,提出对应的算法,并用一个实例证明算法的可行性及适用性。 As an effective model for data processing, the three-way decision has been widely applied to gran-ular computing, concept lattices, and fuzzy mathematical theory since it was proposed. In particular, the combination with granular computing is a practical method for dealing with big data problems. In this paper, a new composite function is established by triangular fuzzy numbers, and the composite data is used to transform the text-type data appearing in the formal context. Then, using the ideas of the three-way decisions and thresholds in fuzzy mathematics, the transformed interval can be effectively enlarged or reduced. Finally, after repeated divisions, an interval division suitable for the actual situation is obtained. According to the textual data partitioning model proposed in this paper, a corresponding algorithm is proposed, and an example is used to prove the feasibility and applicability of the algorithm.
展开▼
机译:三支决策作为一种数据处理的有效模型,自提出以来,被广泛扩展应用到粒计算、概念格及模糊数学理论。特别是与粒计算的结合是处理大数据问题的一种实用方法。本文通过三角模糊数建立一个新的复合函数,并通过这个复合函数转化形式背景中出现的文本型数据。然后,利用三支决策的思想以及模糊数学中的阈值,可以将转化后的区间有效地放大或者缩小。最后,经过反复的划分,得到一种适合实际情况的区间划分。根据本文提出的文本型数据的划分模型,提出对应的算法,并用一个实例证明算法的可行性及适用性。 As an effective model for data processing, the three-way decision has been widely applied to gran-ular computing, concept lattices, and fuzzy mathematical theory since it was proposed. In particular, the combination with granular computing is a practical method for dealing with big data problems. In this paper, a new composite function is established by triangular fuzzy numbers, and the composite data is used to transform the text-type data appearing in the formal context. Then, using the ideas of the three-way decisions and thresholds in fuzzy mathematics, the transformed interval can be effectively enlarged or reduced. Finally, after repeated divisions, an interval division suitable for the actual situation is obtained. According to the textual data partitioning model proposed in this paper, a corresponding algorithm is proposed, and an example is used to prove the feasibility and applicability of the algorithm.
展开▼