project structure

• data
  • raw_data
    • bson
      • products.bson (download the .bson file at this address : https://static.openfoodfacts.org/data/openfoodfacts-mongodbdump.tar.gz)
  • raw_data_mutate (sample in Rdata, made from raw_data_transformation.R)
  • raw_data_transformation.R (script to transform .bson data)
  • raw_data_transformation_personnalnote.R (another script, not used)
• example (other work)
• rsconnect (needed for web deployment)
• server.R
• ui.R
• ui_1.R
• ui_2.R

aim

Give insight for the bureau administration of Openfoodfacts.

technology & method

R & R Shiny. The raw data is a .bson drop (NOSQL). Use of an asynchrone method : generation of a .rds file from .bson to be read in R Shiny.

main variable used

• creator > id creator
• countries_lc > id country
• scans_n > tally scan
• created_t > date
• categories_hierarchy > categorie tag product >>> need to unnest
• nutrition_grades > nutrition grade product >>> need to be unnest

roadmap

• (1) data acquisition > library(mongolite)::find()
• (2) data reduction > library(tidyverse)
• (3) r shiny design & structure > library(shiny)
• (4) r shiny deployment dev (shinyapps.io server)

to do

(1) data acquisition

• increase step by step the sample (currently 10k lines)
• might have scallable issue

(2) data reduction

• better understanding of the data scheme
• unnest and better actions on data
• increase of the number of variable for better insight

(3) r shiny design & structure

• skeleton
• implementation shinydasboard theme
• right organisation for a better understanding
• geographical reading (leaflet)

Third party applications

If you use this SDK, feel free to open a PR to add your application in this list.

Authors