房地产自动评估系统的研究与实现

发布时间：2018-05-24 06:18

本文选题：房地产自动评估 + 网络数据采集　；参考：《华南理工大学》2014年硕士论文

【摘要】：日益活跃的房地产交易市场从各方面影响我国经济，面对活跃的房地产住宅交易市场，房地产评估机构、银行和税务机构需要对住宅价格进行大面积监控，面对如此大量的住宅交易，对于房地产评估机构，要求更快、更准确地完成评估工作；对于银行业，需要对房地产交易市场价格波动进行严密监控，实时调整房地产相关业务政策；对于地方政府，则提供了广阔的税源。房地产价格特征法能从大量不完全的、有噪声的案例数据中，提取出蕴含在其中的信息和知识，是一种面向应用的技术。针对城市大量案例数据从微观入手，通过案例数据建立和分析多个细分的价格模型，通过模型反应真实的细分市场情况。通过小区评级和建立比价关系等措施补充价格模型无法覆盖的区域，全面提升房地产住宅自动评估的覆盖面和准确性。本文首先介绍了关于房地产住宅自动评估所涉及的相关原理，并重点描述了房地产特征因素与特征价格法，讲述网络数据采集的过程，对案例数据进行预处理，介绍特征因素属性标准化规则，介绍小区评级关系和比价关系在批量评估中的作用，描述特征价格模型分析过程。通过网络采集数据的方式极大限度的减少数据来源的资金投入，缩短数据清洗入库的时间周期，比传统的外业调查或购买数据有较好的经济性。通过价格特征法建立的价格模型，避免估价人员主观上的错误判断，，更科学、客观的反映市场的实际情况。结合小区评级和建立小区比价关系，建立多套特征价格模型，细分市场，从微观到宏观反映市场真实价格水平。最后讲述如何验证特征价格模型的反映的市场真实性，通过模型的结果与网络案例价格进行对比，针对误差率作出不同的调整，结合小区评级关系和比价关系提高自动评估的城市覆盖面。
[Abstract]:The increasingly active real estate trading market affects our economy from various aspects. In the face of the active real estate housing trading market, real estate evaluation agencies, banks and tax authorities need to monitor housing prices in a large area. In the face of such a large number of residential transactions, for real estate appraisal agencies, it is necessary to complete the evaluation work faster and more accurately. For the banking industry, it is necessary to monitor closely the price fluctuations in the real estate trading market. Real-time adjustment of real-estate related business policies; for local governments, it provides a broad source of tax. The real estate price feature method can extract information and knowledge from a large number of incomplete and noisy case data. It is an application-oriented technology. Aiming at a large number of urban case data, this paper starts with microcosmic, establishes and analyzes several subdivision price models through case data, and reflects the real market segmentation through the model. In order to improve the coverage and accuracy of the automatic evaluation of real estate housing, we can supplement the areas that can not be covered by the price model through the measures of district rating and establishing price-comparing relationship. This paper first introduces the related principles of the automatic evaluation of real estate housing, and describes the real estate characteristic factors and feature pricing method, describes the process of network data collection, and preprocessing the case data. This paper introduces the standardized rules of attribute of feature factors, introduces the function of the relationship of rating and price in batch evaluation, and describes the analysis process of feature price model. The method of collecting data through the network greatly reduces the investment of the data source and shortens the time period of data cleaning and storage. It is more economical than the traditional field survey or data purchase. The price model established by the price characteristic method can avoid the subjective misjudgment of the valuers and reflect the actual situation of the market more scientifically and objectively. Based on the relationship between community rating and district price comparison, a multi-set of characteristic price model is established to segment the market and reflect the real price level of the market from micro to macro. Finally, how to verify the market authenticity of the feature price model, and compare the results of the model with the network case price, and make different adjustments to the error rate. Combination of residential rating relationship and price relationship to improve the coverage of the automatic assessment of the city.
【学位授予单位】：华南理工大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TP311.52

【参考文献】