基于概率神经网络的建筑学专业本科生创新能力评价  

Innovative Ability Evaluation of Architecture Undergraduate Students based on the Probabilistic Neural Network

作　　者：巩新枝 吴农[2] 熊英 刘娜[1] 姚斌[1] Gong Xinzhi;Wu Nong;Xiong Ying;Liu Na;Yao Bin(School of Civil and Architectural Engineering,Guilin University of Technology,Guilin Guangxi 541004,China;School of Mechanics,Civil Engineering and Architecture,Northwestern Polytechnical University,Xi'an Shaanxi 710072,China;School of Continuing Education,Guilin University of Technology,Guilin Guangxi 541004,China)

机构地区：[1]桂林理工大学土木与建筑工程学院,广西桂林541004 [2]西北工业大学力学与土木建筑学院,陕西西安710072 [3]桂林理工大学继续教育学院,广西桂林541004

出　　处：《城市建筑》2020年第24期84-85,90,共3页Urbanism and Architecture

基　　金：广西高等教育本科教学改革工程一般A类项目(2018JGA176);桂林理工大学本科教学建设与改革新工科专业升级改造项目(xgkz201910)。

摘　　要：培养和提升学生的创新能力是当前高校建筑学本科教育教学改革的着力点。学生创新能力培养提升的影响因素较多且相互作用,用主观思维评判难以得出客观结论。通过设定合理的评价指标,基于概率神经网络的结构和学习规则特性,利用追踪样本数据进行训练,构建建筑学专业本科生创新能力综合评价模型,进行仿真评价实践及验证,进而探讨该方法的实用性与可靠性。Cultivation and promotion of students'innovation ability is one of the key points in the reform of architecture undergraduate education.There are many factors influencing the cultivation and improvement of the students'innovation ability that interact with each other.It is difficult to draw an objective conclusion by subjective thinking.This paper set reasonable evaluation indexes and establish the comprehensive evaluation model of innovation ability of architecture undergraduates based on the probabilistic neural network and the sample data training.The practice and verification of simulation evaluation are carried out in order to explore the practicability and reliability of this method.

关 键 词：概率神经网络 创新能力 评价 建筑学 本科生 

分 类 号：TU18[建筑科学—建筑理论]