Menu
Spatial Data Analysis
City of Melbourne Analysis with QGIS, SQL and Python
This project analyzes the location of train stations in Melbourne CBD and check what businesses are close by the area using nearest neighbors algorithm.
The datasets used for this project are collected from City of Melbourne Data.
The City of Melbourne is an open-source website that provide extensive amount of data for practices.
This project will observe a few of the datasets as follow:
- Municipal Boundary / Melbourne CBD area
- Metro train stations
- Bars and Pubs with Patron Capacity
The datasets used for this project are collected from City of Melbourne Data.
The City of Melbourne is an open-source website that provide extensive amount of data for practices.
This project will observe a few of the datasets as follow:
- Municipal Boundary / Melbourne CBD area
- Metro train stations
- Bars and Pubs with Patron Capacity
Datasets Explanatory
1. Municipal Boundary / Melbourne CBD area.
It is a shapefile dataset that shows the boundary of the City of Melbourne. There are 8 attributes with only 1 feature.
1. Municipal Boundary / Melbourne CBD area.
It is a shapefile dataset that shows the boundary of the City of Melbourne. There are 8 attributes with only 1 feature.
2. Metro Train Stations.
This dataset contains locations of train stations and their accessibility information, such as hearing aid information and lift. There are 4 attributes with 219 features.
This dataset contains locations of train stations and their accessibility information, such as hearing aid information and lift. There are 4 attributes with 219 features.
3. Bars and Pubs with Patron Capacity.
This dataset shows business establishments with its trading name, bar/tavern/pub patron capacity, location and CLUE block and small area designation. There are 10 attributes with 3,863 features.
This dataset shows business establishments with its trading name, bar/tavern/pub patron capacity, location and CLUE block and small area designation. There are 10 attributes with 3,863 features.
Analysis
First step is applying the Municipal Boundary map as the foundation, then analyze which Metro train stations are located within the Melbourne CBD area. After that, filter the train stations that have Lift facility available.
First step is applying the Municipal Boundary map as the foundation, then analyze which Metro train stations are located within the Melbourne CBD area. After that, filter the train stations that have Lift facility available.
The above table shows that there are 6 train stations with lift facility available in the Melbourne CBD area.
The next step is to add another dataset named ‘Bars and Pubs with Patron Capacity’, then perform data cleansing on the data before implementing Nearest Neighbors technique to discover the nearest distance between bars and train stations.
The next step is to add another dataset named ‘Bars and Pubs with Patron Capacity’, then perform data cleansing on the data before implementing Nearest Neighbors technique to discover the nearest distance between bars and train stations.
Data cleansing was performed to remove any blank value in the dataset, before implementing the Machine Learning technique to discover distance between bars and train stations.
The Nearest Neighbors algorithm helped to identify nearest cafes to its train stations and group it as clusters
The Nearest Neighbors algorithm helped to identify nearest cafes to its train stations and group it as clusters
Conclusion
We analysed the train stations that are located within the city area and eliminated the rest of train stations that are out of the boundary. The Richmond train station was found in between the border. Given the scenario that many commuters are transiting in this station, it is accepted as part of the CBD area.
Throughout the analysis, Flag Staff, Flinders Street, Melbourne Central, North Melbourne, Parliament, and Southern Cross are the train stations with Lift facility available.
This project shows an example of how spatial data analysis can improve efficiency in decision making for an organization. The are many techniques that can be implemented, depending on the intended objectives and goals itself.
We analysed the train stations that are located within the city area and eliminated the rest of train stations that are out of the boundary. The Richmond train station was found in between the border. Given the scenario that many commuters are transiting in this station, it is accepted as part of the CBD area.
Throughout the analysis, Flag Staff, Flinders Street, Melbourne Central, North Melbourne, Parliament, and Southern Cross are the train stations with Lift facility available.
This project shows an example of how spatial data analysis can improve efficiency in decision making for an organization. The are many techniques that can be implemented, depending on the intended objectives and goals itself.