Parse Names

If a name is missing or incorrect you can create a pull request

Installation

$ npm i @datagica/parse-names

Usage

Example:

await parse('did you watch romeo et juliette?')

Output:

    [
      {
        "ngram": "romeo",
        "value": {
          "name": "romeo",
          "gender": [
            "m"
          ],
          "culture": [
            "italian"
          ],
          "id": "romeo"
        },
        "score": 1,
        "position": {
          "sentence": 0,
          "word": 3,
          "begin": 14,
          "end": 19
        }
      },
      {
        "ngram": "juliette",
        "value": {
          "name": "juliette",
          "gender": [
            "f"
          ],
          "culture": [
            "french"
          ],
          "id": "juliette"
        },
        "score": 1,
        "position": {
          "sentence": 0,
          "word": 5,
          "begin": 23,
          "end": 31
        }
      }
    ]

TODO

• FIX julián: there is a conflict with "julian" when both keys are in the dataset

• FIX names by searching for � and replace by the corresponding entry in prenoms.csv

• some name has been disabled because they look like common words for instance 'read' (search for comments in the database.js, symbols / or *) in the future we might try to still match them, using a database of last names

• Add more names, using data from U.S. Census Bureau http://deron.meranda.us/data/