comparison_model: Create Comparison Model
In nlanderson9/languagePredictR: Predict Outcomes from Natural Language

Description Usage Arguments Value Examples

This function creates a regression model using a single numeric variable as a predictor, and a specified variable as the outcome. It is intended for comparison against models that use language as a predictor (created by language_model).

1	comparison_model(input, outcome, outcomeType, predictor, progressBar = TRUE)

`input`	A dataframe containing a column with predictor data (numeric variable) and an outcome variable (numeric or two-level factor)
`outcome`	A string consisting of the column name for the outcome variable in `inputDataframe`
`outcomeType`	A string consisting of the type of outcome variable being used - options are "binary" or "continuous"
`predictor`	A string consisting of the column name for the predictor data in `inputDataframe`
`progressBar`	Show a progress bar. Defaults to TRUE.

An object of the type "compModel"

## Not run: 
movie_review_data1$cleanText = clean_text(movie_review_data1$text)

# Using language to predict "Positive" vs. "Negative" reviews
movie_model_valence_language = language_model(movie_review_data1,
                                     outcome = "valence",
                                     outcomeType = "binary",
                                     text = "cleanText")

summary(movie_model_valence_language)

# Is it possible that people write more for negative reviews?
# How does that compare to the language predictors?
movie_review_data1$word_count = corpus(movie_model_data1$cleanText) %>% tokens() %>% ntoken()

# Using word count to predict "Positive" vs. "Negative" reviews
movie_model_valence_wordcount = comparison_model(movie_review_data1,
                                     outcome = "valence",
                                     outcomeType = "binary",
                                     predictor = "word_count")

summary(movie_model_valence_wordcount)

## End(Not run)