fagan2888 / geoquery-project

Pipelines to receive the best area to place a bar

Project definition

The idea for this project is to show the best location where setting up a bar/restaurant whose objective is to serve as a nexus between different companies. After-work culture is gaining relevance and even becoming an essential part in every business. Sacred liquids (also known as beer) allow humans to socialize and often create synergies, businesses and even new ventures.

Criterion to optimize the location:

• Funded companies. They pay income to their employees (instead of equity) that can be spent at our restaurant/bar.
• Companies with at least 10 employees but less than 300. Making business with a very large companies is not very easy if your startup is not big enough. I cannot see SpaceX buying a bunch of nanosatelites to a recent startup. Basically the young startup will not be able to produce the amount, neither deliver on time.
• Sometimes there is someone who ends up feeling too sacred (gets too drunk). It might be interesting not having so many news agencies around to prevent bad PR. Otherwise, we cannot eliminate these companies since they might be extremly helpful for businesses when creating awareness. Therefore, let's affect it slightly negative to the final score.
• Using Google's API, find metro/bus station. It is important our customers can go home safely. Check also for other bars near around. Would be good to have competitors (it always pushes you to work harder and means it is a good place for this business), but let's try to find not a lot of them near around.
• Select the best location.

Steps:

1. Using MongoDB, create a new database and import the json with the data.
2. Write a first query filtering essential data.
3. Clean and prepare the dataset to work with.
4. Export a json including the geopoint and set a geoquery using MongoDB.
5. Print a map showing all the companies we filtered before.
6. Do some querys using Google's API.
7. Create 2 pipelines to: a) Pipeline1: Process the data file and prepare it to convert into a geojson with a geoindex. b) Pipeline2: After geoindexing the file using MongoDB Compass, process this new file and find out the best location.

Files

My-code: Folder with jupyter notebooks files writting the project.

Source: Folder containing two different pipelines:

a) Process the data file and prepare it to convert into a geojson with a geoindex.

b) Process this new file and find out the best area to place our bar. It includes cleaning, geoquerys to google's API, scoring system and finnally plots an interactive map in your browser with the best area.

We only need to charge the dataset with the same structure as the ones we can find in the data file, once we receive the new document insert the geoindex and we will receive the automated result.

My solution

If it ever helps anyone with the location of a bar, feel free to reach me out and invite me to some beers ;)

Links & Resources

• https://docs.mongodb.com/manual/geospatial-queries/
• https://developers.google.com/maps/documentation/geocoding/intro
• https://data.crunchbase.com/docs
• https://developers.google.com/places/web-service/search
• https://www.youtube.com/watch?v=PtV-ZnwCjT0
• https://developers.google.com/places/web-service/search

Next steps

• Optimize pipelines. There are functions that could be written in a more generic way.
• Argparse.
• Autogenerate a pdf file with an automated report.