Lifelong Property Price Prediction: A Case Study for the Toronto Real Estate Market

被引:12
|
作者
Peng, Hao [1 ]
Li, Jianxin [2 ]
Wang, Zheng [3 ]
Yang, Renyu [3 ]
Liu, Mingsheng [4 ]
Zhang, Mingming [5 ]
Yu, Philip S. [6 ]
He, Lifang [7 ]
机构
[1] Beihang Univ, Beijing Adv Innovat Ctr Big Data & Brain Comp, Sch Cyber Sci & Technol, Beijing 100191, Peoples R China
[2] Beihang Univ, Beijing Adv Innovat Ctr Big Data & Brain Comp, Beijing 100191, Peoples R China
[3] Univ Leeds, Sch Comp, Leeds LS2 9JT, England
[4] Shijiazhuang Inst Railway Technol, Shijiazhuang 050041, Peoples R China
[5] UrBrain Technol, Toronto, ON, Canada
[6] Univ Illinois, Dept Comp Sci, Chicago, IL 60607 USA
[7] Lehigh Univ, Dept Comp Sci & Engn, Bethlehem, PA 18015 USA
关键词
Cost accounting; Long short term memory; Space heating; Data models; Water heating; Urban areas; Resistance heating; Heterogeneous information network; graph neural network; LSTM; lifelong learning; house price prediction; NETWORK; DEPENDENCE;
D O I
10.1109/TKDE.2021.3112749
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We present Luce, the first life-long predictive model for automated property valuation. Luce addresses two critical issues of property valuation: the lack of recent sold prices and the sparsity of house data. It is designed to operate on a limited volume of recent house transaction data. As a departure from prior work, Luce organizes the house data in a heterogeneous information network (HIN) where graph nodes are house entities and attributes that are important for house price valuation. We employ a Graph Convolutional Network (GCN) to extract the spatial information from the HIN for house-related data like geographical locations, and then use a Long Short Term Memory (LSTM) network to model the temporal dependencies for house transaction data over time. Unlike prior work, Luce can make effective use of the limited house transactions data in the past few months to update valuation information for all house entities within the HIN. By providing a complete and up-to-date house valuation dataset, Luce thus massively simplifies the downstream valuation task for the targeting properties. We demonstrate the benefit of Luce by applying it to large, real-life datasets obtained from the Toronto real estate market. Extensive experimental results show that Luce not only significantly outperforms prior property valuation methods but also often reaches and sometimes exceeds the valuation accuracy given by independent experts when using the actual realization price as the ground truth.
引用
收藏
页码:2765 / 2780
页数:16
相关论文
共 50 条
  • [41] Real Estate Insights The clash of politics and economics in the UK property market-the case of leaseholds
    Seagraves, Philip
    JOURNAL OF PROPERTY INVESTMENT & FINANCE, 2023, 41 (06) : 629 - 635
  • [42] AN ALGORITHM FOR SETTING THE LIST PRICE OF A RESIDENTIAL REAL-ESTATE PROPERTY
    BANDYOPADHYAY, JK
    BANDYOPADHYAY, B
    INTERFACES, 1981, 11 (04) : 55 - 61
  • [43] Does Property Transaction Matter in the Price Discovery of Real Estate Markets?
    Cheung, William Mingyan
    Lei, James Chicheong
    Tsang, Desmond
    INTERNATIONAL REAL ESTATE REVIEW, 2016, 19 (01): : 27 - 49
  • [44] Distance in Geographic and Characteristics Space for Real Estate Price Prediction
    Ozhegov, Evgeniy M.
    Ozhegova, Alina
    Mitrokhina, Ekaterina
    ANALYSIS OF IMAGES, SOCIAL NETWORKS AND TEXTS, AIST 2019, 2019, 11832 : 63 - 72
  • [45] REAL-ESTATE PRICE PREDICTION UNDER ASYMMETRIC LOSS
    CAIN, M
    JANSSEN, C
    ANNALS OF THE INSTITUTE OF STATISTICAL MATHEMATICS, 1995, 47 (03) : 401 - 414
  • [46] APPLICATION OF FUZZY REGRESSION MODEL FOR REAL ESTATE PRICE PREDICTION
    Sarip, Abdul Ghani
    Hafez, Muhammad Burhan
    Daud, Md. Nasir
    MALAYSIAN JOURNAL OF COMPUTER SCIENCE, 2016, 29 (01) : 15 - 27
  • [47] A SVR BASED FORECASTING APPROACH FOR REAL ESTATE PRICE PREDICTION
    Li, Da-Ying
    Xu, Wei
    Zhao, Hong
    Chen, Rong-Qiu
    PROCEEDINGS OF 2009 INTERNATIONAL CONFERENCE ON MACHINE LEARNING AND CYBERNETICS, VOLS 1-6, 2009, : 970 - +
  • [48] A Hybrid Approach to Real Estate Price Definition: A Case Study in Western Switzerland
    Baillif, Maurin
    de Lapparent, Matthieu
    Kazagli, Evanthia
    REVUE ECONOMIQUE, 2021, 72 (06): : 1055 - 1077
  • [49] The empirical study of the impact of the development credit scale of real estate on the price of real estate
    Cheng, LW
    Tu, C
    Cao, Z
    Proceedings of 2005 International Conference on Construction & Real Estate Management, Vols 1 and 2: CHALLENGE OF INNOVATION IN CONSTRUCTION AND REAL ESTATE, 2005, : 960 - 963
  • [50] The Establishment of The real Estate Price Prediction Model Based on The grey-markov chain: A Case Study of Qingdao City
    Yan, Chun
    Liu, Xinmin
    Liu, Wei
    Sun, Juan
    ADVANCED COMPOSITE MATERIALS, PTS 1-3, 2012, 482-484 : 717 - +