Skip to content

Segmentation of the Statistics Canada’s Set of Proximity Measures – A Clustering Algorithm Approach. UBC MDS capstone project for Statistics Canada.

License

Notifications You must be signed in to change notification settings

sahaavi/Segmentation-of-Proximity-Measures

 
 

Repository files navigation

Proximity Measures Segmentation

A clustering methodology needs to be developed to group continuous proximity measures produced by Statistics Canada.

Team Members

  • Jonah: I like pizza pops.
  • Noman: hmu for sum fire beats: https://www.youtube.com/@Yesuuh 🔥
  • Ricky: Dreams of applying my data science knowledge and skills to practical urban planning research and improvements, and towards building sustainable ‘circle economy’-based societies.
  • Avishek: Motivated Master's of Data Science student with a keen interest in applying my statistical analysis and machine learning skills to drive business development and innovation in the tech industry.

Topic/Interest Description

Even in a world that is rapidly shifting toward digital technologies, evidence has shown that physical proximity between entities, social and economic actors, or consumers and providers of a service, remains a relevant driver of socioeconomic outcomes. Physical proximity to services and amenities has a determinant contribution to economic performances of businesses, quality of life for individuals, and location decisions for people and businesses alike.

About this Project

The Data Exploration and Integration Lab (DEIL) at Statistics Canada is creating a new set of granular measures that assess local access to a variety of amenities such as libraries, parks, educational facilities, hospitals, and more. The measures are generated out of a gravity model approach, where the cost of transportation is weighted by the magnitude or importance of the entity they travel to. The measures are calculated at the dissemination block level in Canada. That is, around 500,000 unique geo-located units. The challenge for the UBC Capstone students will be to implement a clustering algorithm methodology for the segmentation of the measures.

Dataset Description

  • Proximity measures dataset --> Continuous numerical proximity scores for each of the 10 services/amenities, for each dissemination block in Canada.
  • Index of Remoteness --> Continuous numeric remoteness score for each census subdivision in Canada.

Acknowledgements and references

Data Exploration and Integration Lab (DEIL), Statistics Canada

About

Segmentation of the Statistics Canada’s Set of Proximity Measures – A Clustering Algorithm Approach. UBC MDS capstone project for Statistics Canada.

Topics

Resources

License

Code of conduct

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • R 59.6%
  • TeX 40.4%