Gourmet Analytics

TASK: Use these data sets to develop a model that can predict the outcome of the column Like, which indicates whether a customer likes a recipe (your model predicts Like=1) or dislikes it (your model predicts Like=0).

For reviews with a TestSetId and for which you are not given the Like (the “private test set”), you must make predictions that will form your submission. See the “submission_template.csv” for details about the required format. Your project will be evaluated and graded based on these predictions.

Files

.
├── Template
│   ├── crisp_dm.png
│   ├── example_crisp_dm_pipeline.ipynb
│   ├── iris_dataset.csv
│   └── iris_prediction.csv
├── Data
│   ├── Data Description.pdf
│   ├── diet.csv
│   ├── recipes.csv
│   ├── requests.csv
│   ├── reviews.csv
│   └── submission_random.csv

The data

diet.csv

Column	Description
AuthorId	(String, primary key) A unique identifier for the customer.
Diet	(String) The diet option of a user.
Age	(Numeric) The age of the corresponding user.

recipes.csv

Column	Description
RecipeId	(Numeric, primary key) The unique identifier of a recipe.
Name	(String) The name of the recipe.
CookTime	(Numeric) The time it needs to cook the recipe.
PrepTime	(Numeric) The time it needs to prepare the recipe.
RecipeCategory	(String) A label indicating the type of the recipe.
RecipeIngredientQuantities	(String) A label that represents the number of ingredients in the recipe.
RecipeIngredientParts	(String) A label that contains the ingredients in the recipe.
Calories
FatContent
SaturatedFatContent
CarbohydrateContent
FiberContent
SugarContent
ProteinContent	(Numeric) The amount of Macronutrients in kcal/ g per 100g.
SodiumContent
CholesterolContent	(Numeric) Nutrition facts in mg per 100g
RecipeServings	(Numeric) The number of servings that result from this recipe.
RecipeYield	(String) The number of servings that result from this recipe.

reviews.csv

Column	Description
AuthorId
RecipeId	(String, foreign key) The unique identifier of a customer and a recipe.
Rating	(Numeric) A rating between 1 and 5 that the users could submit to evaluate the difficulty of the recipe.
Like	(Numeric) A label that indicates whether a user likes a recipe (=1) or not (=0).
TestSetId	(Numeric) The unique identifier in the test set.

requests.csv

Column	Description
AuthorId
RecipeId	(String, Foreign keys) The unique identifiers of a user and a recipe.
Time	(Numeric) The duration a recipe should take at most (including the time reserved for the preparation and cooking).
HighCalories	(String) A flag that indicates whether the resulting meal/ beverage should have a high number of calories.
LowFat	(Numkleric) A flag that indicates whether the resulting meal/ beverage should include a low amount of fat.
HighProtein	(String) A flag indicating whether the resulting meal/ beverage has a high amount of proteins.
LowSugar	(String) A flag indicates whether the resulting meal/beverage has low sugar.
HighFiber	(Numeric) A flag indicates whether the resulting meal/beverage has high fiber.

yangezheng / gourmetanalytics Goto Github PK