5_project-V-Final-project_foody

Things to do

• Add the count of the ratings and the avg rating to the df_2 dataframe
• List of ingredients that should be always at home (salt, pepper, flour, eggs, etc)
• get the best rating as a result, as a recomendation

Ingredients

• Should be displayed directly at the User interface for selection and fast typing (search botton for faster selection)

Day-to-Day Process

• Dataframe: Cleanig (EDA)
  • Tags
  • Nutrition
  • Steps
  • Ingredients --> needs to be extracted from the DataFrame: create another DataFrame with the unique values of all the ingredients
  • Raiting of each dish
• Recipe Matching Algorithm
  • Functions that allow you to filter by each ingrdient
  • ¿? Partial Matches / Common substitutions
  • ¿? Common variations algorithm (Tomatoe --> Tomatoes)
• Display the results
  • Ingredients to be selected
  • seperated by food category (Dairy, vegetables, meat, gluten, etc)

Idea:

• Insetad of using the dataset to show the information of the recipe, use the webpage, as names are simply the name + recipe_id

To do List for Wednesday

• Create a new data Frame of the count of the ratings and calculate the avg
  • Append and merge these to the Df_2
• EDA the Data Frame and delete anythign not relevant
• change column names if needed
• create ingredients data frame
  • Group ingredients in different categories
• Start working on creating the filters
  • How do I want to filter for the different Dairies / Gluten Free / etc

To do list for Thursday

• Ingredients --> Get only the most relevant ones
  • maybe take the first 300
  • Create a dictionary that will find and equal the values
  • Create categories of these ingredients
• tags --> Need to be sorted: too many
  • take the first 100 maybe and use the most relevant tags
• steps --> Put that as well for each recipe

To do list for Friday

Tableau

• Bins of minutes per ingredients

  • if conditions if the ingredients short medium long

• Counts de color por minutes

• Pie chart de restricciones /Probar radial

• Table/graph de top Rating and Number of Reviews (filter by restrictions) buscar accion - URL markdown

• filtro rank