Developing Rules for Rental Subsidy: An Empirical Housing Study in Taiwan

Homeownership rates have declined, underscoring the significant challenges our society faces in affording homes. This research aims to develop an effective and precise tool for swiftly evaluating and filtering out unqualified applications and establish consistent review criteria for cities and townships across Taiwan. The proposed approach involves the creation of a tool that utilizes particle swarm optimization-based fuzzy hyperrectangular composite neural networks. This paper, chosen without political bias and based on a randomly selected year, uses a data set of 36,086 entries from across Taiwan, with each application containing 10 distinct features for further analysis. The result achieves an impressive accuracy rate of 98.6% and produces 66 recommended rules for determining eligibility for rental subsidies. The contributions of this study are twofold: (1) the rapid auditing tool benefits both government agencies and applicants, streamlining the application process; and (2) the 66 rules generated by the tool offer valuable guidance to internal auditors, expediting audits and reducing personal biases. This promotes a more standardized and efficient workflow.