Dataset for Vacant Land Valuation Using Machine Learning Methods and a New Variable Weighted Quality Score.
Description
This dataset supports research on vacant land valuation by integrating professional judgement, the Analytic Hierarchy Process (AHP), a variable-weighted quality score approach, and machine-learning methods. The data were collected in Thailand from two participant groups: 52 property valuation experts and 324 valuers. The first group provided pairwise-comparison judgements used to determine the relative importance of four main valuation criteria and 15 subcriteria. The second group assessed the quality levels of the same vacant-land characteristics using a structured scoring scale. The dataset includes anonymised questionnaire responses, AHP pairwise-comparison data, factor weights, consistency measures, quality-scoring criteria, processed machine-learning data, model predictions, and performance metrics. The land-related attributes cover physical characteristics, location, legal factors, and external factors, including land area, land shape, frontage width, road conditions, accessibility, zoning, access rights, development potential, utilities, environmental conditions, marketability, and future growth trends. The geographical context relates primarily to vacant land in the central business district of Bangkok, Thailand. Four machine-learning algorithms were applied to the processed data: Decision Tree, Random Forest, Artificial Neural Network, and Gradient Boosted Tree. The associated model outputs include accuracy, precision, recall, and F1-score. These data can be reused to reproduce AHP calculations, examine alternative factor-weighting and multi-criteria decision-making methods, compare professional valuation judgements, develop alternative quality-scoring systems, and test additional machine-learning algorithms for land valuation. The dataset may also support comparative studies across different property markets, provided that differences in local market conditions, legal frameworks, and valuation practices are considered.
Files
Institutions
- Mahasarakham UniversityMaha Sarakham, Maha Sarakham