Prepare a data.table of queries and variables for the FMAT.

Usage

FMAT_query(
  query = "Text with [MASK], optionally with {TARGET} and/or {ATTRIB}.",
  MASK = .(),
  TARGET = .(),
  ATTRIB = .()
)

Arguments

query

Query text (should be a character string/vector with at least one [MASK] token). Multiple queries share the same set of MASK, TARGET, and ATTRIB. For multiple queries with different MASK, TARGET, and/or ATTRIB, please use FMAT_query_bind to combine them.

MASK

A named list of [MASK] target words. Must be single words in the vocabulary of a certain masked language model.

For model vocabulary, see, e.g., https://huggingface.co/bert-base-uncased/raw/main/vocab.txt

Infrequent words may be not included in a model's vocabulary, and in this case you may insert the words into the context by specifying either TARGET or ATTRIB.

TARGET, ATTRIB

A named list of Target/Attribute words or phrases. If specified, then query must contain {TARGET} and/or {ATTRIB} (in all uppercase and in braces) to be replaced by the words/phrases.

Value

A data.table of queries and variables.

Examples

FMAT_query("[MASK] is a nurse.", MASK = .(Male="He", Female="She"))
#>                 query   MASK M_pair M_word
#>                <fctr> <fctr> <fctr> <fctr>
#> 1: [MASK] is a nurse.   Male      1     He
#> 2: [MASK] is a nurse. Female      1    She

FMAT_query(
  c("[MASK] is {TARGET}.", "[MASK] works as {TARGET}."),
  MASK = .(Male="He", Female="She"),
  TARGET = .(Occupation=c("a doctor", "a nurse", "an artist"))
)
#>        qid                     query   MASK M_pair M_word     TARGET
#>     <fctr>                    <fctr> <fctr> <fctr> <fctr>     <fctr>
#>  1:      1       [MASK] is {TARGET}.   Male      1     He Occupation
#>  2:      1       [MASK] is {TARGET}. Female      1    She Occupation
#>  3:      1       [MASK] is {TARGET}.   Male      1     He Occupation
#>  4:      1       [MASK] is {TARGET}. Female      1    She Occupation
#>  5:      1       [MASK] is {TARGET}.   Male      1     He Occupation
#>  6:      1       [MASK] is {TARGET}. Female      1    She Occupation
#>  7:      2 [MASK] works as {TARGET}.   Male      1     He Occupation
#>  8:      2 [MASK] works as {TARGET}. Female      1    She Occupation
#>  9:      2 [MASK] works as {TARGET}.   Male      1     He Occupation
#> 10:      2 [MASK] works as {TARGET}. Female      1    She Occupation
#> 11:      2 [MASK] works as {TARGET}.   Male      1     He Occupation
#> 12:      2 [MASK] works as {TARGET}. Female      1    She Occupation
#>           T_pair    T_word
#>           <fctr>    <fctr>
#>  1: Occupation.1  a doctor
#>  2: Occupation.1  a doctor
#>  3: Occupation.2   a nurse
#>  4: Occupation.2   a nurse
#>  5: Occupation.3 an artist
#>  6: Occupation.3 an artist
#>  7: Occupation.1  a doctor
#>  8: Occupation.1  a doctor
#>  9: Occupation.2   a nurse
#> 10: Occupation.2   a nurse
#> 11: Occupation.3 an artist
#> 12: Occupation.3 an artist

FMAT_query(
  "The [MASK] {ATTRIB}.",
  MASK = .(Male=c("man", "boy"),
           Female=c("woman", "girl")),
  ATTRIB = .(Masc=c("is masculine", "has a masculine personality"),
             Femi=c("is feminine", "has a feminine personality"))
)
#>                    query   MASK M_pair M_word ATTRIB      A_pair
#>                   <fctr> <fctr> <fctr> <fctr> <fctr>      <fctr>
#>  1: The [MASK] {ATTRIB}.   Male      1    man   Masc Masc-Femi.1
#>  2: The [MASK] {ATTRIB}.   Male      2    boy   Masc Masc-Femi.1
#>  3: The [MASK] {ATTRIB}. Female      1  woman   Masc Masc-Femi.1
#>  4: The [MASK] {ATTRIB}. Female      2   girl   Masc Masc-Femi.1
#>  5: The [MASK] {ATTRIB}.   Male      1    man   Masc Masc-Femi.2
#>  6: The [MASK] {ATTRIB}.   Male      2    boy   Masc Masc-Femi.2
#>  7: The [MASK] {ATTRIB}. Female      1  woman   Masc Masc-Femi.2
#>  8: The [MASK] {ATTRIB}. Female      2   girl   Masc Masc-Femi.2
#>  9: The [MASK] {ATTRIB}.   Male      1    man   Femi Masc-Femi.1
#> 10: The [MASK] {ATTRIB}.   Male      2    boy   Femi Masc-Femi.1
#> 11: The [MASK] {ATTRIB}. Female      1  woman   Femi Masc-Femi.1
#> 12: The [MASK] {ATTRIB}. Female      2   girl   Femi Masc-Femi.1
#> 13: The [MASK] {ATTRIB}.   Male      1    man   Femi Masc-Femi.2
#> 14: The [MASK] {ATTRIB}.   Male      2    boy   Femi Masc-Femi.2
#> 15: The [MASK] {ATTRIB}. Female      1  woman   Femi Masc-Femi.2
#> 16: The [MASK] {ATTRIB}. Female      2   girl   Femi Masc-Femi.2
#>                          A_word
#>                          <fctr>
#>  1:                is masculine
#>  2:                is masculine
#>  3:                is masculine
#>  4:                is masculine
#>  5: has a masculine personality
#>  6: has a masculine personality
#>  7: has a masculine personality
#>  8: has a masculine personality
#>  9:                 is feminine
#> 10:                 is feminine
#> 11:                 is feminine
#> 12:                 is feminine
#> 13:  has a feminine personality
#> 14:  has a feminine personality
#> 15:  has a feminine personality
#> 16:  has a feminine personality

Prepare a data.table of queries and variables for the FMAT.

Usage

Arguments

Value

See also

Examples