keyword_relevance: Watson Natural Language Understanding: Relevance of Keywords

In johannesharmse/watsonNLU:

Description

See the sign-up documentation for step by step instructions to secure your own username and password to enable you to use the Watson NLU API. The keyword_relevance function takes a text or URL input, along with the input type. The function then returns a dataframe that contains keywords and their likelihood of being a keyword, from the given input. See the keyword_relevance documentation for more usage cases.

Usage

keyword_relevance(input = NULL, input_type = NULL, limit = NULL,
  version = "?version=2018-03-16")

Arguments

input	Either a text string input or website URL. Either text or url argument has to be specified, but not both.
input_type	Specify what type of input was entered. Either text or url argument has to be specified, but not both.
limit	The number of keywords to return.
version	The release date of the API version to use. Default value is version="?version=2018-03-16"

Value

A dataframe containing a list of keywords and their corresponding likelihoods for the given input.

See Also

keyword_sentiment, \link[watsonNLU]{text_categories}, \link[watsonNLU]{keyword_emotions}, \link[watsonNLU]{auth_NLU}

Examples

credentials <- readRDS("../tests/testthat/credentials.rds")
username <- credentials$username
password <- credentials$password

# Authenticate using Watson NLU API Credentials
auth_NLU(username, password)
# Authenticate using Watson NLU API Credentials
auth_NLU(username="XXXX", password="XXXX")

# Top 5 keywords from the text input.
keyword_relevance(input = 'This is a great API wrapper', input_type='text', limit = 5)

# Top 5 keywords from the URL input.
keyword_relevance(input = 'http://www.nytimes.com/guides/well/how-to-be-happy', input_type='url', limit = 5)

johannesharmse/watsonNLU documentation built on May 21, 2019, 12:04 p.m.