TY - GEN
T1 - AffordAD
T2 - 14th Annual Undergraduate Research Conference on "ICT for Resilient and Sustainable Infrastructure", URC 2022
AU - Almusalami, Aisha
AU - Habuza, Tetiana
AU - Singh, Harsh
AU - Zaki, Nazar
N1 - Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Improving housing affordability is one of the United nations' sustainable development goals (SDGs). It is also highlighted in Abu Dhabi Community Facility Planning Standards (PS5 affordable housing planning standard). The proposed work presents a user-friendly tool to estimate affordability. First, a user selects a location of interest. The tool automatically finds the district where the selected geocoordinate is located. Then, it counts different facilities within 1 km of range and number of past accidents happened in the district corresponding to selected point. A user can manually select the type of the unit (studio, villa, and apartment), number of rooms and alter the auto-filled fields to see their effect on housing affordability. It provides flexibility in city planning. All facilities within 1km are displayed on the map for more informative and data-driven decisions. Then a user can click the predict button to infer through a trained machine learning model. This model is developed using data from 42 heterogeneous datasets retrieved from Abu Dhabi open data platform. Data from different sources such as GIS, facilities, and accidents were pre-processed and integrated to develop the solution. A user can estimate the affordability of a particular location based on its surrounding area. Additionally, detailed reports about the district and nearby facilities can be downloaded for further analysis. Our solution will help users to find the best location to live in based on their budget. It will help decisionmakers to evaluate housing affordability in the future. Our machine learning model (Gradient Boosting) that is behind the tool presents a high performance in terms of Mean Absolute Error (MAE) and Coefficient of Determination (R-Squared).
AB - Improving housing affordability is one of the United nations' sustainable development goals (SDGs). It is also highlighted in Abu Dhabi Community Facility Planning Standards (PS5 affordable housing planning standard). The proposed work presents a user-friendly tool to estimate affordability. First, a user selects a location of interest. The tool automatically finds the district where the selected geocoordinate is located. Then, it counts different facilities within 1 km of range and number of past accidents happened in the district corresponding to selected point. A user can manually select the type of the unit (studio, villa, and apartment), number of rooms and alter the auto-filled fields to see their effect on housing affordability. It provides flexibility in city planning. All facilities within 1km are displayed on the map for more informative and data-driven decisions. Then a user can click the predict button to infer through a trained machine learning model. This model is developed using data from 42 heterogeneous datasets retrieved from Abu Dhabi open data platform. Data from different sources such as GIS, facilities, and accidents were pre-processed and integrated to develop the solution. A user can estimate the affordability of a particular location based on its surrounding area. Additionally, detailed reports about the district and nearby facilities can be downloaded for further analysis. Our solution will help users to find the best location to live in based on their budget. It will help decisionmakers to evaluate housing affordability in the future. Our machine learning model (Gradient Boosting) that is behind the tool presents a high performance in terms of Mean Absolute Error (MAE) and Coefficient of Determination (R-Squared).
KW - Affordability
KW - Gradient Boosting
KW - PS5 Affordable housing planning standard
KW - SDG
UR - http://www.scopus.com/inward/record.url?scp=85149932281&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85149932281&partnerID=8YFLogxK
U2 - 10.1109/URC58160.2022.10054228
DO - 10.1109/URC58160.2022.10054228
M3 - Conference contribution
AN - SCOPUS:85149932281
T3 - Proceedings of the 2022 14th Annual Undergraduate Research Conference on "ICT for Resilient and Sustainable Infrastructure", URC 2022
BT - Proceedings of the 2022 14th Annual Undergraduate Research Conference on "ICT for Resilient and Sustainable Infrastructure", URC 2022
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 23 November 2022 through 24 November 2022
ER -
