Package 'aws.comprehend'

aws.comprehend

Description

AWS Comprehend Client Package

Details

Client for AWS Comprehend (https://aws.amazon.com/comprehend0, a cloud natural language processing service that can perform a number of quantitative text analyses, including language detection, sentiment analysis, and feature extraction.

Author(s)

Thomas J. Leeper <[email protected]>

See Also

detect_language, detect_sentiment, detect_entities, detect_phrases

---

Bind and index a ResultList

Description

Turn a list of data.frames (of different lengths and potentially empty) into a single indexed data.frame. Useful to process a ResultList from 'comprehendHTTP'.

Usage

bind_and_index(index, df_list)

Arguments

index	Vector of indices
df_list	List of data.frames to bind and index. Should NOT be a data.frame.

Details

'index' and 'df_list' should be the same length. An error is raised otherwise.

bind_and_index(1:2, list(data.frame(col = "a"), data.frame(col = "b")))

bind_and_index(1:3, list( data.frame(col = "a"), data.frame(), data.frame(c("b", "c"))))

---

Execute AWS Comprehend API Request

Description

This is the workhorse function to execute calls to the Comprehend API.

Usage

comprehendHTTP(
  action,
  query = list(),
  headers = list(),
  body = NULL,
  verbose = getOption("verbose", FALSE),
  region = Sys.getenv("AWS_DEFAULT_REGION", "us-east-1"),
  key = NULL,
  secret = NULL,
  session_token = NULL,
  service = c("comprehend", "comprehendmedical"),
  ...
)

Arguments

action	A character string specifying the API action to take
query	An optional named list containing query string parameters and their character values.
headers	A list of headers to pass to the HTTP request.
body	A request body
verbose	A logical indicating whether to be verbose. Default is given by options("verbose").
region	A character string containing the AWS region. If missing, defaults to “us-east-1”.
key	A character string containing an AWS Access Key ID. See locate_credentials.
secret	A character string containing an AWS Secret Access Key. See locate_credentials.
session_token	A character string containing an AWS Session Token. See locate_credentials.
service	the Comprehend service to use. Currently either 'comprehend' for the base service or 'comprehendmedical' for the Comprehend Medical service.
...	Additional arguments passed to GET.

Details

This function constructs and signs an Polly API request and returns the results thereof, or relevant debugging information in the case of error.

Value

If successful, a named list. Otherwise, a data structure of class “aws-error” containing any error message(s) from AWS and information about the request attempt.

Author(s)

Thomas J. Leeper

---

Detect named entities in a source text

Description

Detect entities in a source text

Usage

detect_entities(text, language = "en", ...)

Arguments

text	A character string containing a text to entities analyze, or a character vector to perform analysis separately for each element.
language	A character string containing a two-letter language code. Currently “en” and “es” are supported.
...	Additional arguments passed to comprehendHTTP.

Value

A data frame

Examples

## Not run: 
  # simple example
  detect_entities("Amazon provides web services. Jeff is their leader.")
  
  txt <-c("Amazon provides web services, like Google.",
          "Jeff is their leader.")
  detect_entities(txt)

## End(Not run)

---

Detect language in a source text

Description

Detect language(s) in a source text

Usage

detect_language(text, ...)

Arguments

text	A character string containing a textual source, or a character vector to detect languages separately for each element.
...	Additional arguments passed to comprehendHTTP.

Value

A data frame of language probabilities.

Examples

## Not run: 
  # simple example
  detect_language("This is a test sentence in English")
  
  # two languages in a single text
  txt <- "A: ¡Hola! ¿Como está, usted?\nB: Ça va bien. Merci. Et toi?"
  detect_language(txt)

  # "batch" mode
  detect_language(c("A: ¡Hola! ¿Como está, usted?",
                    "B: Ça va bien. Merci. Et toi?"))

## End(Not run)

---

Detect named entities in a source medical text

Description

Detect entities in a source medical text

Usage

detect_medical_entities(text, language = "en", version = c("2", "1"), ...)

Arguments

text	A character string containing a text to entities analyze, or a character vector to perform analysis separately for each element.
language	A character string containing a two-letter language code. Currently only “en” is supported.
version	A character string containing the version of the API that should be used. Currently only "1" or "2" are supported.
...	Additional arguments passed to comprehendHTTP.

Value

A data frame

Examples

## Not run: 
  # simple example
  medical_detect_entities("Mrs. Smith comes in today complaining of shortness of breath.")
  
  txt <-c("Mrs. Smith comes in today.",
          "She is complaining of shortnesss of breath.")
  medical_detect_entities(txt)

## End(Not run)

---

Detect Protected Health Information (PHI) in a source medical text

Description

Detect Protected Health Information (PHI) in a source medical text

Usage

detect_medical_phi(text, language = "en", ...)

Arguments

text	A character string containing a text to entities analyze, or a character vector to perform analysis separately for each element.
language	A character string containing a two-letter language code. Currently only “en” is supported.
...	Additional arguments passed to comprehendHTTP.

Value

A data frame

Examples

## Not run: 
  # simple example
  medical_detect_phi("Mrs. Smith comes in today complaining of shortness of breath.")
  
  txt <-c("Mrs. Smith comes in today.",
          "She is complaining of shortnesss of breath.")
  medical_detect_phi(txt)

## End(Not run)

---

Detect key phrases

Description

Detect key phrases in a source text

Usage

detect_phrases(text, language = "en", ...)

Arguments

text	A character string containing a text to analyze, or a character vector to perform analysis separately for each element.
language	A character string containing a two-letter language code. Currently “en” and “es” are supported.
...	Additional arguments passed to comprehendHTTP.

Value

A data frame

Examples

## Not run: 
  # simple example
  detect_phrases("Amazon provides web services. Jeff is their leader.")
  
  txt <-c("Amazon provides web services.",
          "Jeff is their leader.")
  detect_phrases(txt)

## End(Not run)

---

Detect sentiment in a source text

Description

Detect sentiment in a source text

Usage

detect_sentiment(text, language = "en", ...)

Arguments

text	A character string containing a text to sentiment analyze, or a character vector to perform analysis separately for each element.
language	A character string containing a two-letter language code. Currently “en” and “es” are supported.
...	Additional arguments passed to comprehendHTTP.

Value

A data frame

Examples

## Not run: 
  # simple example
  detect_sentiment("I have never been happier. This is the best day ever.")

  txt <-c("I have never been happier. This is the best day ever.",
          "I have always been happier. This is the worst day ever.")
  detect_sentiment(txt)

## End(Not run)

---

Detect syntax in a source text

Description

Detect syntax in a source text

Usage

detect_syntax(text, language = "en", ...)

Arguments

text	A character string containing a text to syntax analyze, or a character vector to perform analysis separately for each element.
language	A character string containing a two-letter language code.
...	Additional arguments passed to comprehendHTTP.

Value

A data frame

Examples

## Not run: 
  # simple example
  detect_syntax("The quick brown fox jumps over the lazy dog.")

  txt <-c("The quick brown fox jumps over the lazy dog.",
          "I have never been happier!")
  detect_syntax(txt)

## End(Not run)

---

Flatten embedded data.frames (1 level max)

Description

Flatten embedded data.frames (1 level max)

Usage

flatten(df)

Arguments

df	data.frame to flatten

---