Fraczek, Witold, ESRI, Redlands, Lange, Jian, ESRI, Redlands, Carsten Lange, California State Polytechnic University, Pomona
The presentation elaborates on findings from a collaboration project between GIS professionals and data scientists to apply machine learning algorithms to predict urban development in a study area known as the Research Triangle in North Carolina. The predictive model used in the project is the Random Forest machine learning algorithm, a popular supervised learning algorithm based on a multiple of decision trees. The study applies Esri’s advanced GIS software ArcGIS Pro, various R packages in Rstudio, and the R-ArcGIS Bridge, which is an open source R package from Esri that allows the passing of data between ArcGIS Pro and R. Factors that affect urban development such as the proximity to roads, urban centers, environmental protected areas, flood zones, as well as terrain characteristics and projected population growth are considered in the predictive analysis model. The goal of the project is to identify locations with a high probability of urban development in the study area. The prototype project shows promising prediction results. As population is growing, new areas need to be converted from their current land use types into urban land use. Which areas would be most suitable for urbanization? How probable is urban development in a specific area? The answers to these questions are critical for government agencies such as planning departments in need of better understanding of urban growth in order to make better policies. It is also beneficial for private investors who are searching for locations to make profitable investments. Investors looking for opportunities to invest in real estate and commercial infrastructure such as shops or restaurants, can use the model to locate suitable areas.
The presentation will also share the project’s collaboration experience on how to integrate knowledge and skills from GIS and Data Science fields. The availability of large dataset, advanced spatial analysis, and machine learning tools allows data scientists to combine technologies to take predictive analytics to the next level. Consequently, existing research about predicting urban development often requires working with more than one software. A key factor that makes the smooth workflow is the open source R-ArcGIS Bridge package that allows R to dynamically access ArcGIS data, and save R results back to ArcGIS datasets.
