Data-driven Adaptive Personalized Property Investment Risk Analysis: Frameworks, Methods and System

Publication Type:
Issue Date:
Full metadata record
The risk analysis for real estate property investment which incurred a high cost and high risk has been qualitatively and quantitatively assessed by various techniques. These techniques consider the heuristic risk factors mainly based on the expert survey, weigh and rank the factors using algorithms and mathematical formulas and decide the best investment based on the performance index of alternatives given. Currently, identifying, weighing, and ranking of the risk factors in investment decisions largely depend on expert judgment using traditional approaches. Additionally, current research is lacking in considering investors’ different preferences and requirements. The motivation of this study is how to identify, weigh and rank the risk analysis factors when experts have a different point of view or judgments’ and expert subjective justifications. This thesis describes a new personalized multidimensional process (PMP) framework based on knowledge discovery to overcome the weaknesses of existing risk identification and measurement techniques. This framework comprises two new methods namely personalized association mapping (PAM) method and personalized multidimensional sensitivity analysis (PM-SA) method. This thesis also proposes an adaptive personalized property investment risk analysis (APPIRA) method to identify the property investment determinants. This APPIRA method adopts a data-driven and personalization technique to weigh risk factors and ranking using a multi-criteria decision-making model by applying the TOPSIS method for optimal solutions. The existing real estate analytic systems, which only serve as search tools and do not benefit homebuyers in terms of search time, flexibility, and intuitive results. Thus, a prototype of adaptive personalized property investment risk analysis system (APPIRAS) developed to validate the framework and methods proposed to overcome the limitations. The innovations of this research were the justification of risk factor identification as the determinant for different objectives; weighing which is based on historical data-driven to decision support using the knowledge discovery approach (analytical decision-making) and the investor’s adaptive personalization of the risk factors which fulfil their requirements; and ranking using multi-criteria decision-making model. The methodology in this research incorporated literature review, frameworks development, methods development, prototype software system development and evaluation. A case study has been developed to show the applicability of the developed framework, methods, and system. The outcomes of this study have a significant impact in helping investors to achieve their objectives and APPIRAS as the decision support tools to achieve optimal decisions. These methods can be applied to practice and benefit the property industry directly.
Please use this identifier to cite or link to this item: