针对参变量数据元集的可拓建筑策划分类知识挖掘方法  

Methods for Extension Architectural Programming Classification Knowledge Mining on Parametric Data Set

作　　者：邹广天[1,2] 张斯[1,2] 郭强[1,2] 丁俐娟[1,2] 

机构地区：[1]哈尔滨工业大学建筑学院 [2]哈尔滨工业大学建筑计划与设计研究所,哈尔滨150006

出　　处：《科技导报》2014年第36期43-47,共5页Science & Technology Review

基　　金：国家自然科学基金项目(51178132)

摘　　要：为解决可拓分类知识挖掘方法难以满足可拓建筑策划数据及问题的特殊性需求,将可集合成参变量数据元集的可拓建筑策划数据作为研究对象,提出适用于可拓建筑策划的可拓分类知识挖掘流程与方法,即数据的预处理、选取评价特征并确定权重、建立关联函数并划分区间、计算获取可拓分类知识。拓展可拓建筑策划数据挖掘在方法层面的研究,以方便获取重要的可拓建筑策划分类知识,为可拓建筑策划提供决策支持,为计算机辅助建筑策划提供新思路。Extension architectural programming classification knowledge mining on parametric data set is an application of extension data mining in the extension architectural programming area. However, the existing methods for classification knowledge mining are difficult to meet the particularity demands of extension architectural programming. The paper takes extension architectural programming data which can be set into parametric data set as a research object, and puts forward a procedure and certain methods for classification knowledge mining suitable for extension architectural programming, including pretreatment of data, selecting evaluation characteristics and determining the weights, establishing the correlation function and dividing the interval, and calculating and accessing the extension classification knowledge. The paper aims to expand the method-level research on extension architectural programming data mining to facilitate access to important extension architectural programming classification knowledge, and provides a new idea for computer-aided architectural programming.

关 键 词：可拓建筑策划 可拓数据挖掘 可拓分类知识 参变量数据元集 

分 类 号：TU201.4[建筑科学—建筑设计及理论]