Identification of locational influence on real property values using data mining methods
Bibliographic record
Abstract
The value of real estate is an important matter for municipal authorities, since property tax is one of their main budget sources. Its estimation tends to be a complex process, owing to the diversity of factors affecting it. One of those factors is property location, which embraces the geographic relationship between the property and the surrounding local amenities. Hedonic modelling is frequently applied to estimate the value of a property; to consider the influence of property location within such models, the region under analysis is usually divided into homogeneous areas. This division can introduce a bias (a particular vision) related to the modifiable areal unit problem. Our intent in this paper is to apply data mining techniques to address a possible valuer bias, a particular valuer’s vision, in the current City of Calgary assessment model. Employing the decision tree technique, one locational attribute (Sub-Neighbourhood) was represented by the (x, y) coordinates of the properties, with approximately 96% correct classification with respect to their City of Calgary sub-neighbourhood designation. By adopting the regression tree technique, we show that it is possible to explain approximately 73% variability of the Sale Price attribute, using only the attribute Sub-Neighbourhood or the (x, y) coordinates as input. In general, the results showed a consistent relationship between property value and location. Additionally, the sale price patterns of actual properties do not conform strictly to the politico-administrative units adopted by the city. Those patterns usually cross the unit boundaries limits or are mixed inside a unit. Our results suggest that using a property’s spatial coordinates, instead of political-administrative subdivisions, to express its location, would lead to more accurate results and not incur the possibility of bias.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.013 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.006 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.002 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".