pcparts_app:

Web scraping PCPartpicker.com for value-based recommendations

The finished app can be found here. A full write up of this project can be accessed via this link.

The contents of this main folder are the ipython EDA notebook, the final presentation slides, the webscraping folder and the final part recommendation app.

Altogether 5 programming languages were used for this project (in order of chronology):

1. JavaScript (PhantomJS) - webscraping tool, used to grab all of our required data
2. Bash - command line scripting to automate and coordinate python, PhantomJS scripts
3. Python - BeautifulSoup parsing of scraped data; initial EDA and data preparation in ipython notebook
4. SQL - local database created to store all of the cleaned data; most of the app logic is constructed in SQL
5. R - the final app is constructed with Shiny and hosted on shinyapps.io