【优秀硕博论文】基于人工神经网络的乌克兰民居房地产评估研究.docVIP

管理学硕士学位论文Artificial Neural Network Application for Ukrainian Real Estate Market Appraisal. 学科、专业 ： 论文主审人 ： 哈尔滨工程大学 201年月 摘 要 由于房地产评估本身具有现实性、市场性等特性，传统的评估方法过于依赖评估人员的经验，主观随意性大，难以保证评估结果的客观、科学与公正性。针对民居房地产价格与其影响因素之间的复杂、不确定性和非线性关系特性，本文将人工神经网络引入乌克兰民居房地产评估中，对传统房地产评估方法进行改进，以减少由于评估人员主观因素造成的评估结果偏差。本文在前人的文献的基础上，主要做了以下几个方面的工作： ①.综合分析了乌克兰民居房产评估的迫切性和必要性，调研了民居房地产评估的研究背景和国内外研究现状，明确当前乌克兰民居房地产评估存在的问题和研究目的、意义。 ②.综合分析了民居房地产评估的原则和神经网络分析方法，并结合乌克兰民居房地产存在的问题来分析神经网络评估特性，进而综合分析这些评估方法中存在的一些问题。 ③.结合乌克兰民居房地产评估现状，构建了基于神经网络的房地产评估模型，对相应的模型数据进行准备，再将数据应用于模型。 ④.对评估模型进行了实证分析。先引入实证对象的相关信息和特性，在此基础上展示相关的计算过程，并对计算结果进行了综合分析。 关键词：Abstract With the rapid development of the Ukrainian economy, and real estate transactions have become increasingly active, especially the significant increase in residential real estate transactions, related to residential real estate appraisal line also will get a rapid development. The traditional method is to real estate appraisal real estate appraisal staff appraisal in accordance with certain procedures, the analysis of the factors affecting the real estate prices, based on experience with their own assessment of the price of real estate made estimates and judgments. As the real estate appraisal itself realistic, market and other characteristics, too dependent on the traditional appraisal methods experienced evaluators, subjective and arbitrary, it is difficult to ensure the results of the appraisal objective, scientific and fair. For residential real estate prices and their impact complex factors, uncertainties and nonlinear relationship characteristics, the paper introduced ANN Ukraine residential real estate appraisal, real estate appraisal traditional methods improved to reduce the appraisal subjective factors assessment bias. Based on previous literature, mainly to do the following work areas: ①.Comprehensive analysis of the urgency and necessity of Ukrainian residential real estate appraisal, research the background research status and residential real estate appr