Implement a parser for the Recipe1M dataset

[WagnerMarcel]

GanttStart: 2022-01-14 GanttDue: 2022-02-01

[mscholl96]

Only use recipes with no more than 20 ingredients and 30 instructions

Output data format:

[
  {
    "title": "NULL",
    "ingredients": [
      {
        "ingredient": "NULL",
        "amount": "NULL",
        "unit":"NULL"
        "instruction": "NULL" (optional, left out in first approach)
      }
    ],
   "instructions": [
    {
      "instruction":"NULL"
    }
   ]
  }
]

Formatted with: https://jsonformatter.org

[SchatziHub]

Open points:

normalization of ingredients

• singularize from pattern lib
• lower case everything
• replace " by inch:
• split e.g. "salt and pepper"
• prefilter more invalid stuff:
• normalize/adjust amounts of "half" ingredients:
• cluster/fuse ingredients that are very similar (e.g. salt/kosher salt, butter/unsalted butter)

[SchatziHub]

bug of pattern.singularize: flour turns into flmy