Predicting residential building age from map data

The age of a building influences its form and fabric composition and this in turn is critical to inferring its energy performance. However, often this data is unknown. In this paper, we present a methodology to automatically identify the construction period of houses, for the purpose of urban energy...

Full description

Bibliographic Details
Main Authors: Rosser, Julian F., Boyd, Doreen S., Long, G., Zakhary, Sameh, Mao, Y., Robinson, D.
Format: Article
Language:English
Published: Elsevier 2018
Online Access:https://eprints.nottingham.ac.uk/54994/
_version_ 1848799095659954176
author Rosser, Julian F.
Boyd, Doreen S.
Long, G.
Zakhary, Sameh
Mao, Y.
Robinson, D.
author_facet Rosser, Julian F.
Boyd, Doreen S.
Long, G.
Zakhary, Sameh
Mao, Y.
Robinson, D.
author_sort Rosser, Julian F.
building Nottingham Research Data Repository
collection Online Access
description The age of a building influences its form and fabric composition and this in turn is critical to inferring its energy performance. However, often this data is unknown. In this paper, we present a methodology to automatically identify the construction period of houses, for the purpose of urban energy modelling and simulation. We describe two major stages to achieving this – a per-building classification model and post-classification analysis to improve the accuracy of the class inferences. In the first stage, we extract measures of the morphology and neighbourhood characteristics from readily available topographic mapping, a high-resolution Digital Surface Model and statistical boundary data. These measures are then used as features within a random forest classifier to infer an age category for each building. We evaluate various predictive model combinations based on scenarios of available data, evaluating these using 5-fold cross-validation to train and tune the classifier hyper-parameters based on a sample of city properties. A separate sample estimated the best performing cross-validated model as achieving 77% accuracy. In the second stage, we improve the inferred per-building age classification (for a spatially contiguous neighbourhood test sample) through aggregating prediction probabilities using different methods of spatial reasoning. We report on three methods for achieving this based on adjacency relations, near neighbour graph analysis and graph-cuts label optimisation. We show that post-processing can improve the accuracy by up to 8 percentage points.
first_indexed 2025-11-14T20:30:13Z
format Article
id nottingham-54994
institution University of Nottingham Malaysia Campus
institution_category Local University
language English
last_indexed 2025-11-14T20:30:13Z
publishDate 2018
publisher Elsevier
recordtype eprints
repository_type Digital Repository
spelling nottingham-549942020-03-13T04:30:12Z https://eprints.nottingham.ac.uk/54994/ Predicting residential building age from map data Rosser, Julian F. Boyd, Doreen S. Long, G. Zakhary, Sameh Mao, Y. Robinson, D. The age of a building influences its form and fabric composition and this in turn is critical to inferring its energy performance. However, often this data is unknown. In this paper, we present a methodology to automatically identify the construction period of houses, for the purpose of urban energy modelling and simulation. We describe two major stages to achieving this – a per-building classification model and post-classification analysis to improve the accuracy of the class inferences. In the first stage, we extract measures of the morphology and neighbourhood characteristics from readily available topographic mapping, a high-resolution Digital Surface Model and statistical boundary data. These measures are then used as features within a random forest classifier to infer an age category for each building. We evaluate various predictive model combinations based on scenarios of available data, evaluating these using 5-fold cross-validation to train and tune the classifier hyper-parameters based on a sample of city properties. A separate sample estimated the best performing cross-validated model as achieving 77% accuracy. In the second stage, we improve the inferred per-building age classification (for a spatially contiguous neighbourhood test sample) through aggregating prediction probabilities using different methods of spatial reasoning. We report on three methods for achieving this based on adjacency relations, near neighbour graph analysis and graph-cuts label optimisation. We show that post-processing can improve the accuracy by up to 8 percentage points. Elsevier 2018-09-13 Article PeerReviewed application/pdf en cc_by_nc_nd https://eprints.nottingham.ac.uk/54994/1/Estimating_building_age_submission_Aug2018_accepted_pre-typesetting.pdf Rosser, Julian F., Boyd, Doreen S., Long, G., Zakhary, Sameh, Mao, Y. and Robinson, D. (2018) Predicting residential building age from map data. Computers, Environment and Urban Systems . ISSN 0198-9715 https://www.sciencedirect.com/science/article/pii/S0198971518300851?via%3Dihub doi:10.1016/j.compenvurbsys.2018.08.004 doi:10.1016/j.compenvurbsys.2018.08.004
spellingShingle Rosser, Julian F.
Boyd, Doreen S.
Long, G.
Zakhary, Sameh
Mao, Y.
Robinson, D.
Predicting residential building age from map data
title Predicting residential building age from map data
title_full Predicting residential building age from map data
title_fullStr Predicting residential building age from map data
title_full_unstemmed Predicting residential building age from map data
title_short Predicting residential building age from map data
title_sort predicting residential building age from map data
url https://eprints.nottingham.ac.uk/54994/
https://eprints.nottingham.ac.uk/54994/
https://eprints.nottingham.ac.uk/54994/