| Title: | Analyse Open-Ended Survey Responses in Finnish |
|---|---|
| Description: | Annotates Finnish textual survey responses into CoNLL-U format using Finnish treebanks from <https://universaldependencies.org/format.html> using UDPipe as described in Straka and Straková (2017) <doi:10.18653/v1/K17-3009>. Formatted data is then analysed using single or comparison n-gram plots, wordclouds, summary tables and Concept Network plots. The Concept Network plots use the TextRank algorithm as outlined in Mihalcea, Rada & Tarau, Paul (2004) <https://aclanthology.org/W04-3252/>. |
| Authors: | Adeline Clarke [cre, aut], Krista Lagus [aut], Katja Laine [aut], Maria Litova [aut], Matti Nelimarkka [aut], Joni Oksanen [aut], Jaakko Peltonen [aut], Tuukka Oikarinen [aut], Jani-Matti Tirkkonen [aut], Ida Toivanen [aut], Maria Valaste [aut], Shannon Emilia Carson [ctb], Sirpa Lappalainen [ctb], Tuukka Puonti [ctb], Kimmo Vehkalahti [ctb], DARIAH-FI [cph, fnd] |
| Maintainer: | Adeline Clarke <[email protected]> |
| License: | MIT + file LICENSE |
| Version: | 2.1.1 |
| Built: | 2025-03-06 16:27:09 UTC |
| Source: | https://github.com/dariah-fi-survey-concept-network/finnsurveytext |
This data contains background variables and the responses to q3 "Missä asioissa olet hyvä? (Avokysymys)", q7 "Kertoisitko, mitä sinun mielestäsi kiusaaminen on? (Avokysymys)", and q11 "Mikä tekee sinut iloiseksi? (Avokysymys)" in the FSD3134 Lapsibarometri 2016 dataset.
childchild
## 'child' A dataframe with 414 rows and 8 columns:
FSD case id
'Which things are you good at?' response text
'What do you think bullying is?' response text
'What makes you happy?' response text
Weight
Gender)
Major region)
Daycare before pre-school
<https://urn.fi/urn:nbn:fi:fsd:T-FSD3134>
This data contains background variables and the responses to q11_1 'Jatka lausetta: Kehitysmaa on maa, jossa... (Avokysymys)', q11_2 'Jatka lausetta: Kehitysyhteistyö on toimintaa, jossa... (Avokysymys)', q11_3' Jatka lausetta: Maailman kolme suurinta ongelmaa ovat... (Avokysymys)' in the FSD2821 Nuorten ajatuksia kehitysyhteistyöstä 2012 dataset.
dev_coopdev_coop
## 'dev_coop' A dataframe with 925 rows and 9 columns:
FSD case id
response text for q11_1
response text for q11_2
response text for q11_3
Weight
Gender
Year of Birth
Region of Residence
Education level
<https://urn.fi/urn:nbn:fi:fsd:T-FSD2821>
This data contains English text responses to ""Joe’s doctor told him that he would need to return in two weeks to find out whether or not his condition had improved. But when Joe asked the receptionist for an appointment, he was told that it would be over a month before the next available appointment. What should Joe do?" as well as categorisation of these responses by two coders as either destructive, passive, somewhat proactive, or proactive.
english_sample_surveyenglish_sample_survey
## 'english_sample_survey' A dataframe with 585 rows and 5 columns:
ID
Label: destructive, passive, somewhat proactive, or proactive
Label from coder 1
Label from coder 2
Text of response
<https://doi.org/10.7802/2474>
This data contains the responses to q7 "Kertoisitko, mitä sinun mielestäsi kiusaaminen on? (Avokysymys)" in the FSD3134 Lapsibarometri 2016 dataset in CoNLL-U format with NLTK stopwords and punctuation removed plus weights and background variables.
fst_childfst_child
## 'fst_child' A dataframe with 1580 rows and 18 columns:
the identifier of the document
the identifier of the paragraph
the identifier of the sentence
the text of the sentence for which this token is part of
Word index, integer starting at 1 for each new sentence; may be a range for multi-word tokens; may be a decimal number for empty nodes.
Word form or punctuation symbol.
Lemma or stem of word form.
Universal part-of-speech tag.
Language-specific part-of-speech tag; underscore if not available.
List of morphological features from the universal feature inventory or from a defined language-specific extension; underscore if not available.
Head of the current word, which is either a value of token_id or zero (0).
Universal dependency relation to the HEAD (root iff HEAD = 0) or a defined language-specific subtype of one.
Enhanced dependency graph in the form of a list of head-deprel pairs.
Any other annotation.
Weight
Gender
Major region
Daycare before pre-school
<https://urn.fi/urn:nbn:fi:fsd:T-FSD3134>
This data contains the responses to q7 "Kertoisitko, mitä sinun mielestäsi kiusaaminen on? (Avokysymys)" in the FSD3134 Lapsibarometri 2016 dataset in CoNLL-U format with NLTK stopwords and punctuation removed.
fst_child_2fst_child_2
## 'fst_child_2' A dataframe with 1580 rows and 14 columns:
the identifier of the document
the identifier of the paragraph
the identifier of the sentence
the text of the sentence for which this token is part of
Word index, integer starting at 1 for each new sentence; may be a range for multi-word tokens; may be a decimal number for empty nodes.
Word form or punctuation symbol.
Lemma or stem of word form.
Universal part-of-speech tag.
Language-specific part-of-speech tag; underscore if not available.
List of morphological features from the universal feature inventory or from a defined language-specific extension; underscore if not available.
Head of the current word, which is either a value of token_id or zero (0).
Universal dependency relation to the HEAD (root iff HEAD = 0) or a defined language-specific subtype of one.
Enhanced dependency graph in the form of a list of head-deprel pairs.
Any other annotation.
<https://urn.fi/urn:nbn:fi:fsd:T-FSD3134>
Creates a Concept Network plot from a list of edges and nodes (and their respective weights) which indicates unique words in this plot in comparison to another Network.
fst_cn_compare_plot( edges, nodes, concepts, unique_lemmas, name = NULL, concept_colour = "#cd1719", unique_colour = "#4DAF4A", min_edge = NULL, max_edge = NULL, min_node = NULL, max_node = NULL, title_size = 20 )fst_cn_compare_plot( edges, nodes, concepts, unique_lemmas, name = NULL, concept_colour = "#cd1719", unique_colour = "#4DAF4A", min_edge = NULL, max_edge = NULL, min_node = NULL, max_node = NULL, title_size = 20 )
edges |
Output of ‘fst_cn_edges()', dataframe of ’edges' connecting two words. |
nodes |
Output of 'fst_cn_nodes()', dataframe of relevant lemmas and their associated pagerank. |
concepts |
List of terms which have been searched for, separated by commas. |
unique_lemmas |
List of unique lemmas, output of 'fst_cn_get_unique()' |
name |
An optional "name" for the plot, default is 'NULL' and a generic title ("TextRank extracted keyword occurrences") will be used. |
concept_colour |
Colour to display concept words, default is '"indianred"'. |
unique_colour |
Colour to display unique words, default is '"darkgreen"'. |
min_edge |
A numeric value for the scale of the edges, the smallest co_occurrence value for an edge across all Networks to be plotted together. |
max_edge |
A numeric value for the scale of the edges, the largest co_occurrence value for an edge across all Networks to be plotted together. |
min_node |
A numeric value for the scale of the nodes, the smallest pagerank value for a node across all Networks to be plotted together. |
max_node |
A numeric value for the scale of the nodes, the largest pagerank value for a node across all Networks to be plotted together. |
title_size |
size to display plot title |
Plot of concept network with concept and unique words (nodes) highlighted.
pos_filter <- c("NOUN", "VERB", "ADJ", "ADV") e1 <- fst_cn_edges(fst_child, "lyödä", pos_filter = pos_filter) e2 <- fst_cn_edges(fst_child, "lyöminen", pos_filter = pos_filter) n1 <- fst_cn_nodes(fst_child, e1) n2 <- fst_cn_nodes(fst_child, e2) u <- fst_cn_get_unique_separate(n1, n2) fst_cn_compare_plot(e1, n1, "lyödä", unique_lemma = u) fst_cn_compare_plot(e2, n2, "lyöminen", u, unique_colour = "purple")pos_filter <- c("NOUN", "VERB", "ADJ", "ADV") e1 <- fst_cn_edges(fst_child, "lyödä", pos_filter = pos_filter) e2 <- fst_cn_edges(fst_child, "lyöminen", pos_filter = pos_filter) n1 <- fst_cn_nodes(fst_child, e1) n2 <- fst_cn_nodes(fst_child, e2) u <- fst_cn_get_unique_separate(n1, n2) fst_cn_compare_plot(e1, n1, "lyödä", unique_lemma = u) fst_cn_compare_plot(e2, n2, "lyöminen", u, unique_colour = "purple")
This function takes a string of terms (separated by commas) or a single term and, using 'fst_cn_search()' find words connected to these searched terms. Then, a dataframe is returned of 'edges' between two words which are connected together in an frequently-occurring n-gram containing a concept term.
fst_cn_edges( data, concepts, threshold = NULL, norm = "number_words", pos_filter = NULL )fst_cn_edges( data, concepts, threshold = NULL, norm = "number_words", pos_filter = NULL )
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
concepts |
List of terms to search for, separated by commas. |
threshold |
A minimum number of occurrences threshold for 'edge' between searched term and other word, default is 'NULL'. Note, the threshold is applied before normalisation. |
norm |
The method for normalising the data. Valid settings are '"number_words"' (the number of words in the responses), '"number_resp"' (the number of responses), or 'NULL' (raw count returned, default, also used when weights are applied). |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' to include all UPOS tags. |
Dataframe of co-occurrences between two connected words.
con <- "kiusata, lyöminen" fst_cn_edges(fst_child, con, pos_filter = c("NOUN", "VERB", "ADJ", "ADV")) fst_cn_edges(fst_child, con, pos_filter = 'VERB, NOUN') fst_cn_edges(fst_child, "lyöminen", threshold = 2, norm = "number_resp")con <- "kiusata, lyöminen" fst_cn_edges(fst_child, con, pos_filter = c("NOUN", "VERB", "ADJ", "ADV")) fst_cn_edges(fst_child, con, pos_filter = 'VERB, NOUN') fst_cn_edges(fst_child, "lyöminen", threshold = 2, norm = "number_resp")
Takes at least two tables of nodes and pagerank (output of 'fst_cn_nodes()') and finds nodes unique to one table.
fst_cn_get_unique(list)fst_cn_get_unique(list)
list |
A list of top nodes |
Dataframe of words and whether word is unique or not.
pos_filter <- 'NOUN, VERB, ADJ, ADV' e1 <- fst_cn_edges(fst_child, "lyödä", pos_filter = pos_filter) e2 <- fst_cn_edges(fst_child, "lyöminen", pos_filter = pos_filter) n1 <- fst_cn_nodes(fst_child, e1) n2 <- fst_cn_nodes(fst_child, e2) list_of_nodes <- list() list_of_nodes <- append(list_of_nodes, list(n1)) list_of_nodes <- append(list_of_nodes, list(n2)) fst_cn_get_unique(list_of_nodes)pos_filter <- 'NOUN, VERB, ADJ, ADV' e1 <- fst_cn_edges(fst_child, "lyödä", pos_filter = pos_filter) e2 <- fst_cn_edges(fst_child, "lyöminen", pos_filter = pos_filter) n1 <- fst_cn_nodes(fst_child, e1) n2 <- fst_cn_nodes(fst_child, e2) list_of_nodes <- list() list_of_nodes <- append(list_of_nodes, list(n1)) list_of_nodes <- append(list_of_nodes, list(n2)) fst_cn_get_unique(list_of_nodes)
Takes at least two tables of nodes and pagerank (output of 'fst_cn_nodes()') and finds nodes unique to one table.
fst_cn_get_unique_separate(table1, table2, ...)fst_cn_get_unique_separate(table1, table2, ...)
table1 |
The first table. |
table2 |
The second table. |
... |
Any other tables you want to include. |
Dataframe of words and whether word is unique or not.
pos_filter <- c("NOUN", "VERB", "ADJ", "ADV") e1 <- fst_cn_edges(fst_child, "lyödä", pos_filter = pos_filter) e2 <- fst_cn_edges(fst_child, "lyöminen", pos_filter = pos_filter) n1 <- fst_cn_nodes(fst_child, e1) n2 <- fst_cn_nodes(fst_child, e2) fst_cn_get_unique_separate(n1, n2)pos_filter <- c("NOUN", "VERB", "ADJ", "ADV") e1 <- fst_cn_edges(fst_child, "lyödä", pos_filter = pos_filter) e2 <- fst_cn_edges(fst_child, "lyöminen", pos_filter = pos_filter) n1 <- fst_cn_nodes(fst_child, e1) n2 <- fst_cn_nodes(fst_child, e2) fst_cn_get_unique_separate(n1, n2)
This function takes a string of terms (separated by commas) or a single term and, using 'textrank_keywords()' from 'textrank' package, filters data based on 'pos_filter' ranks words which are the filtered for those connected to search terms.
fst_cn_nodes(data, edges, pos_filter = NULL)fst_cn_nodes(data, edges, pos_filter = NULL)
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
edges |
Output of 'fst_cn_edges()', dataframe of co-occurrences between two words. |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' to include all UPOS tags. |
A dataframe containing relevant lemmas and their associated pagerank.
con <- "kiusata, lyöminen" cb <- fst_child edges <- fst_cn_edges(cb, con, pos_filter = c("NOUN", "VERB", "ADJ", "ADV")) edges2 <- fst_cn_edges(cb, con, pos_filter = 'NOUN, VERB, ADJ, ADV') fst_cn_nodes(cb, edges, c("NOUN", "VERB", "ADJ", "ADV")) fst_cn_nodes(cb, edges, 'NOUN, VERB, ADJ, ADV')con <- "kiusata, lyöminen" cb <- fst_child edges <- fst_cn_edges(cb, con, pos_filter = c("NOUN", "VERB", "ADJ", "ADV")) edges2 <- fst_cn_edges(cb, con, pos_filter = 'NOUN, VERB, ADJ, ADV') fst_cn_nodes(cb, edges, c("NOUN", "VERB", "ADJ", "ADV")) fst_cn_nodes(cb, edges, 'NOUN, VERB, ADJ, ADV')
Creates a Concept Network plot from a list of edges and nodes (and their respective weights).
fst_cn_plot(edges, nodes, concepts, title = NULL)fst_cn_plot(edges, nodes, concepts, title = NULL)
edges |
Output of ‘fst_cn_edges()', dataframe of ’edges' connecting two words. |
nodes |
Output of 'fst_cn_nodes()', dataframe of relevant lemmas and their associated pagerank. |
concepts |
List of terms which have been searched for, separated by commas. |
title |
Optional title for plot, default is 'NULL' and a generic title ("TextRank extracted keyword occurrences") will be used. |
Plot of Concept Network.
con <- "kiusata, lyöminen" cb <- fst_child edges <- fst_cn_edges(cb, con, pos_filter = c("NOUN", "VERB", "ADJ", "ADV")) nodes <- fst_cn_nodes(cb, edges, c("NOUN", "VERB", "ADJ", "ADV")) fst_cn_plot(edges = edges, nodes = nodes, concepts = con)con <- "kiusata, lyöminen" cb <- fst_child edges <- fst_cn_edges(cb, con, pos_filter = c("NOUN", "VERB", "ADJ", "ADV")) nodes <- fst_cn_nodes(cb, edges, c("NOUN", "VERB", "ADJ", "ADV")) fst_cn_plot(edges = edges, nodes = nodes, concepts = con)
This function takes a string of terms (separated by commas) or a single term and, using 'textrank_keywords()' from 'textrank' package, filters data based on 'pos_filter' and finds words connected to search terms.
fst_cn_search(data, concepts, pos_filter = NULL)fst_cn_search(data, concepts, pos_filter = NULL)
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
concepts |
String of terms to search for, separated by commas. |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' to include all UPOS tags. |
Dataframe of n-grams containing searched terms.
con <- "kiusata, lyöminen, lyödä, potkia" pf <- c("NOUN", "VERB", "ADJ", "ADV") pf2 <- "NOUN, VERB, ADJ, ADV" fst_cn_search(fst_child, concepts = con, pos_filter = pf) fst_cn_search(fst_child, concepts = con, pos_filter = pf2) fst_cn_search(fst_child, concepts = con)con <- "kiusata, lyöminen, lyödä, potkia" pf <- c("NOUN", "VERB", "ADJ", "ADV") pf2 <- "NOUN, VERB, ADJ, ADV" fst_cn_search(fst_child, concepts = con, pos_filter = pf) fst_cn_search(fst_child, concepts = con, pos_filter = pf2) fst_cn_search(fst_child, concepts = con)
Creates a comparison wordcloud showing words that occur differently between each group. Data is split based on different values in the 'field' column of formatted data. Results will be shown within the plots pane.
fst_comparison_cloud( data, field, pos_filter = NULL, max = 100, norm = NULL, use_svydesign_weights = FALSE, use_svydesign_field = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE, exclude_nulls = FALSE, rename_nulls = "null_data" )fst_comparison_cloud( data, field, pos_filter = NULL, max = 100, norm = NULL, use_svydesign_weights = FALSE, use_svydesign_field = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE, exclude_nulls = FALSE, rename_nulls = "null_data" )
data |
A dataframe of text in CoNLL-U format with additional 'field' column for splitting data. |
field |
Column in 'data' used for splitting groups |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' which means all word types included. |
max |
The maximum number of words to display, default is '100'. |
norm |
The method for normalising the data. Valid settings are '"number_words"' (the number of words in the responses), '"number_resp"' (the number of responses), or 'NULL' (raw count returned, default, also used when weights are applied). |
use_svydesign_weights |
Option to weight words in the wordcloud using weights from a svydesign object containing the raw data, default is 'FALSE' |
use_svydesign_field |
Option to get 'field' for splitting the data from the svydesign object, default is 'FALSE' |
id |
ID column from raw data, required if 'use_svydesign_weights = TRUE' and must match the 'docid' in formatted 'data'. |
svydesign |
A svydesign object which contains the raw data and weights. |
use_column_weights |
Option to weight words in the wordcloud using weights from formatted data which includes addition 'weight' column, default is 'FALSE' |
exclude_nulls |
Whether to include NULLs in 'field' column, default is 'FALSE' |
rename_nulls |
What to fill NULL values with if 'exclude_nulls = FALSE'. |
A comparison cloud from wordcloud package.
fst_comparison_cloud(fst_child, 'gender', max = 50) s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' c2 <- fst_child_2 fst_comparison_cloud(c2, 'gender', NULL, 100, NULL, TRUE, TRUE, i, s) T <- TRUE fst_comparison_cloud(fst_dev_coop, 'education_level', use_column_weights = T) pf <- c("NOUN", "VERB", "ADJ", "ADV") pf2 <- "NOUN, VERB, ADJ, ADV" fst_comparison_cloud(fst_dev_coop, 'gender', pos_filter = pf) fst_comparison_cloud(fst_dev_coop, 'gender', pos_filter = pf2) fst_comparison_cloud(fst_dev_coop, 'gender', norm = 'number_resp')fst_comparison_cloud(fst_child, 'gender', max = 50) s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' c2 <- fst_child_2 fst_comparison_cloud(c2, 'gender', NULL, 100, NULL, TRUE, TRUE, i, s) T <- TRUE fst_comparison_cloud(fst_dev_coop, 'education_level', use_column_weights = T) pf <- c("NOUN", "VERB", "ADJ", "ADV") pf2 <- "NOUN, VERB, ADJ, ADV" fst_comparison_cloud(fst_dev_coop, 'gender', pos_filter = pf) fst_comparison_cloud(fst_dev_coop, 'gender', pos_filter = pf2) fst_comparison_cloud(fst_dev_coop, 'gender', norm = 'number_resp')
This function takes a string of terms (separated by commas) or a single term and, using 'textrank_keywords()' from 'textrank' package, filters data based on 'pos_filter' and finds words connected to search terms. Then it plots a Concept Network based on the calculated weights of these terms and the frequency of co-occurrences.
fst_concept_network( data, concepts, threshold = NULL, norm = "number_words", pos_filter = NULL, title = NULL )fst_concept_network( data, concepts, threshold = NULL, norm = "number_words", pos_filter = NULL, title = NULL )
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
concepts |
List of terms to search for, separated by commas. |
threshold |
A minimum number of occurrences threshold for 'edge' between searched term and other word, default is 'NULL'. Note, the threshold is applied before normalisation. |
norm |
The method for normalising the data. Valid settings are '"number_words"' (the number of words in the responses), '"number_resp"' (the number of responses), or 'NULL' (raw count returned, default, also used when weights are applied). |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' to include all UPOS tags. |
title |
Optional title for plot, default is 'NULL' and a generic title ("TextRank extracted keyword occurrences") will be used. |
Plot of Concept Network.
data <- fst_child con <- "kiusata, lyöminen" pf <- c("NOUN", "VERB", "ADJ", "ADV") title <- "Bullying Concept Network" fst_concept_network(data, concepts = con, pos_filter = pf, title = title)data <- fst_child con <- "kiusata, lyöminen" pf <- c("NOUN", "VERB", "ADJ", "ADV") title <- "Bullying Concept Network" fst_concept_network(data, concepts = con, pos_filter = pf, title = title)
This function takes a string of terms (separated by commas) or a single term and, using 'textrank_keywords()' from 'textrank' package, filters data based on 'pos_filter' and finds words connected to search terms for each group. Then it plots a Concept Network for each group based on the calculated weights of these terms and the frequency of co-occurrences, indicating any words that are unique to each group's Network plot.
fst_concept_network_compare( data, concepts, field, norm = NULL, threshold = NULL, pos_filter = NULL, use_svydesign_field = FALSE, id = "", svydesign = NULL, exclude_nulls = FALSE, rename_nulls = "null_data", title_size = 20, subtitle_size = 15 )fst_concept_network_compare( data, concepts, field, norm = NULL, threshold = NULL, pos_filter = NULL, use_svydesign_field = FALSE, id = "", svydesign = NULL, exclude_nulls = FALSE, rename_nulls = "null_data", title_size = 20, subtitle_size = 15 )
data |
A dataframe of text in CoNLL-U format with additional 'field' column for splitting data. |
concepts |
List of terms to search for, separated by commas. |
field |
Column in 'data' used for splitting groups |
norm |
The method for normalising the data. Valid settings are '"number_words"' (the number of words in the responses, default), '"number_resp"' (the number of responses), or 'NULL' (raw count returned). |
threshold |
A minimum number of occurrences threshold for 'edge' between searched term and other word, default is 'NULL'. Note, the threshold is applied before normalisation. |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' to include all UPOS tags. |
use_svydesign_field |
Option to get 'field' for splitting the data from a svydesign object, default is 'FALSE' |
id |
ID column from raw data, required if 'use_svydesign_weights = TRUE' and must match the 'docid' in formatted 'data'. |
svydesign |
A svydesign object which contains the raw data and weights. |
exclude_nulls |
Whether to include NULLs in 'field' column, default is 'FALSE' |
rename_nulls |
What to fill NULL values with if 'exclude_nulls = FALSE'. |
title_size |
size to display plot title |
subtitle_size |
size to display title of individual concept network |
Multiple concept network plots with concept and unique words highlighted.
con1 <- "lyödä, lyöminen" fst_concept_network_compare(fst_child, concepts = con1, field = 'gender') s <- survey::svydesign(id=~1, weights= ~paino, data = child) c2 <- fst_child_2 i <- 'fsd_id' fst_concept_network_compare(c2, con1, 'gender', NULL, NULL, NULL, TRUE, i, s) con2 <- "köyhyys, nälänhätä, sota" fst_concept_network_compare(fst_dev_coop, con2, 'gender')con1 <- "lyödä, lyöminen" fst_concept_network_compare(fst_child, concepts = con1, field = 'gender') s <- survey::svydesign(id=~1, weights= ~paino, data = child) c2 <- fst_child_2 i <- 'fsd_id' fst_concept_network_compare(c2, con1, 'gender', NULL, NULL, NULL, TRUE, i, s) con2 <- "köyhyys, nälänhätä, sota" fst_concept_network_compare(fst_dev_coop, con2, 'gender')
This data contains the responses to Development Cooperation q11_3 dataset in CoNLL-U format with NLTK stopwords and punctuation removed plus weights and background variables.
fst_dev_coopfst_dev_coop
## 'fst_dev_coop' A dataframe with 4192 rows and 19 columns:
the identifier of the document
the identifier of the paragraph
the identifier of the sentence
the text of the sentence for which this token is part of
Word index, integer starting at 1 for each new sentence; may be a range for multi-word tokens; may be a decimal number for empty nodes.
Word form or punctuation symbol.
Lemma or stem of word form.
Universal part-of-speech tag.
Language-specific part-of-speech tag; underscore if not available.
List of morphological features from the universal feature inventory or from a defined language-specific extension; underscore if not available.
Head of the current word, which is either a value of token_id or zero (0).
Universal dependency relation to the HEAD (root iff HEAD = 0) or a defined language-specific subtype of one.
Enhanced dependency graph in the form of a list of head-deprel pairs.
Any other annotation.
Weight
Gender
Year of Birth
Region of Residence
<https://urn.fi/urn:nbn:fi:fsd:T-FSD2821>
This data contains the responses to Development Cooperation q11_3 dataset in CoNLL-U format with NLTK stopwords and punctuation removed.
fst_dev_coop_2fst_dev_coop_2
## 'fst_dev_coop_2' A dataframe with 4192 rows and 14 columns:
the identifier of the document
the identifier of the paragraph
the identifier of the sentence
the text of the sentence for which this token is part of
Word index, integer starting at 1 for each new sentence; may be a range for multi-word tokens; may be a decimal number for empty nodes.
Word form or punctuation symbol.
Lemma or stem of word form.
Universal part-of-speech tag.
Language-specific part-of-speech tag; underscore if not available.
List of morphological features from the universal feature inventory or from a defined language-specific extension; underscore if not available.
Head of the current word, which is either a value of token_id or zero (0).
Universal dependency relation to the HEAD (root iff HEAD = 0) or a defined language-specific subtype of one.
Enhanced dependency graph in the form of a list of head-deprel pairs.
Any other annotation.
<https://urn.fi/urn:nbn:fi:fsd:T-FSD2821>
Returns a tibble containing all available stopword lists for the language, their contents, and the size of the lists.
fst_find_stopwords(language = "fi")fst_find_stopwords(language = "fi")
language |
two-letter ISO code of the language for the stopword list |
A tibble containing the stopwords lists.
fst_find_stopwords() fst_find_stopwords(language = 'et')fst_find_stopwords() fst_find_stopwords(language = 'et')
Creates a dataframe in CoNLL-U format from a dataframe containing text from using the [udpipe] package and a language model plus any additional columns that are included such as 'weights' or columns added through 'add_cols'.
fst_format(data, question, id, model = "ftb", weights = NULL, add_cols = NULL)fst_format(data, question, id, model = "ftb", weights = NULL, add_cols = NULL)
data |
A dataframe of survey responses which contains an open-ended question. |
question |
The column in the dataframe which contains the open-ended question. |
id |
The column in the dataframe which contains the ids for the responses. |
model |
A language model available for [udpipe]. '"ftb"' (default) or '"tdt"' are recognised as shorthand for "finnish-ftb" and "finnish-tdt". The full list is available in the [udpipe] documentation or via 'fst_print_available_models()'. |
weights |
Optional, the column of the dataframe which contains the respective weights for each response. |
add_cols |
Optional, a column (or columns) from the dataframe which contain other information you'd like to retain (for instance, covariate columnns for splitting the data for comparison plots). |
Dataframe of annotated text in CoNLL-U format plus any additional columns.
## Not run: i <- "fsd_id" fst_format(data = child, question = "q7", id = i) fst_format(data = child, question = "q7", id = i, model = "tdt") fst_format(data = child, question = "q7", id = i, weights="paino") cols <- c("gender", "major_region", "daycare_before_school") fst_format(child, question = "q7", id = i, add_cols = cols) fst_format(child, question = "q7", id = i, add_cols = "gender, major_region") fst_format(child, question = 'q7', id = i, model = 'swedish-talbanken') unlink("finnish-ftb-ud-2.5-191206.udpipe") unlink("finnish-tdt-ud-2.5-191206.udpipe") unlink("swedish-talkbanken-ud-2.5-191206.udpipe") ## End(Not run)## Not run: i <- "fsd_id" fst_format(data = child, question = "q7", id = i) fst_format(data = child, question = "q7", id = i, model = "tdt") fst_format(data = child, question = "q7", id = i, weights="paino") cols <- c("gender", "major_region", "daycare_before_school") fst_format(child, question = "q7", id = i, add_cols = cols) fst_format(child, question = "q7", id = i, add_cols = "gender, major_region") fst_format(child, question = 'q7', id = i, model = 'swedish-talbanken') unlink("finnish-ftb-ud-2.5-191206.udpipe") unlink("finnish-tdt-ud-2.5-191206.udpipe") unlink("swedish-talkbanken-ud-2.5-191206.udpipe") ## End(Not run)
Creates a dataframe in CoNLL-U format from a 'svydesign' object including text using the [udpipe] package and a language model plus weights if these are included in the 'svydesign' object and any columns added through 'add_cols'.
fst_format_svydesign( svydesign, question, id, model = "ftb", use_weights = TRUE, add_cols = NULL )fst_format_svydesign( svydesign, question, id, model = "ftb", use_weights = TRUE, add_cols = NULL )
svydesign |
A 'svydesign' object which contains an open-ended question. |
question |
The column in the dataframe which contains the open-ended question. |
id |
The column in the dataframe which contains the ids for the responses. |
model |
A language model available for [udpipe]. '"ftb"' (default) or '"tdt"' are recognised as shorthand for "finnish-ftb" and "finnish-tdt". The full list is available in the [udpipe] documentation or via 'fst_print_available_models()'. |
use_weights |
Optional, whether to use weights within the 'svydesign' |
add_cols |
Optional, a column (or columns) from the dataframe which contain other information you'd like to retain (for instance, dimension columnns for splitting the data for comparison plots). |
Dataframe of annotated text in CoNLL-U format plus any additional columns.
## Not run: i <- "fsd_id" svy_child <- survey::svydesign(id=~1, weights= ~paino, data = child) fst_format_svydesign(svy_child, question = 'q7', id = 'fsd_id') fst_format_svydesign(svy_child, question = 'q7', id = i, use_weights = FALSE) cols <- c('gender', 'major_region') fst_format_svydesign(svy_child, 'q7', 'fsd_id', add_cols = cols) svy_dev <- survey::svydesign(id = ~1, weights = ~paino, data = dev_coop) fst_format_svydesign(svy_dev, 'q11_1', 'fsd_id', add_cols = 'gender, region') fst_format_svydesign(svy_dev, 'q11_2', 'fsd_id', 'finnish-ftb') unlink("finnish-ftb-ud-2.5-191206.udpipe") unlink("finnish-tdt-ud-2.5-191206.udpipe") ## End(Not run)## Not run: i <- "fsd_id" svy_child <- survey::svydesign(id=~1, weights= ~paino, data = child) fst_format_svydesign(svy_child, question = 'q7', id = 'fsd_id') fst_format_svydesign(svy_child, question = 'q7', id = i, use_weights = FALSE) cols <- c('gender', 'major_region') fst_format_svydesign(svy_child, 'q7', 'fsd_id', add_cols = cols) svy_dev <- survey::svydesign(id = ~1, weights = ~paino, data = dev_coop) fst_format_svydesign(svy_dev, 'q11_1', 'fsd_id', add_cols = 'gender, region') fst_format_svydesign(svy_dev, 'q11_2', 'fsd_id', 'finnish-ftb') unlink("finnish-ftb-ud-2.5-191206.udpipe") unlink("finnish-tdt-ud-2.5-191206.udpipe") ## End(Not run)
Creates a plot of the most frequently-occurring words (unigrams) within the data. Optionally, weights can be provided either through a 'weight' column in the formatted data, or from a 'svydesign' object with the raw (preformatted) data.
fst_freq( data, number = 10, norm = NULL, pos_filter = NULL, strict = TRUE, name = NULL, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )fst_freq( data, number = 10, norm = NULL, pos_filter = NULL, strict = TRUE, name = NULL, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
number |
The number of top words to return, default is '10'. |
norm |
The method for normalising the data. Valid settings are '"number_words"' (the number of words in the responses, default), '"number_resp"' (the number of responses), or 'NULL' (raw count returned). |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' which means all word types included. |
strict |
Whether to strictly cut-off at 'number' (ties are alphabetically ordered), default is 'TRUE'. |
name |
An optional "name" for the plot to add to title, default is 'NULL'. |
use_svydesign_weights |
Option to weight words in the plot using weights from a 'svydesign' containing the raw data, default is 'FALSE' |
id |
ID column from raw data, required if 'use_svydesign_weights = TRUE' and must match the 'docid' in formatted 'data'. |
svydesign |
A 'svydesign' which contains the raw data and weights, required if 'use_svydesign_weights = TRUE'. |
use_column_weights |
Option to weight words in the plot using weights from formatted data which includes addition 'weight' column, default is 'FALSE' |
Plot of top words.
fst_freq(fst_child, number = 12, norm = 'number_resp', name = "All") fst_freq(fst_child, use_column_weights = TRUE) s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' fst_freq(fst_child_2, use_svydesign_weights = TRUE, svydesign = s, id = i)fst_freq(fst_child, number = 12, norm = 'number_resp', name = "All") fst_freq(fst_child, use_column_weights = TRUE) s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' fst_freq(fst_child_2, use_svydesign_weights = TRUE, svydesign = s, id = i)
Find top and unique top words for different groups of participants. Data is split based on different values in the 'field' column of formatted data. Results will be shown within the plots pane.
fst_freq_compare( data, field, number = 10, norm = NULL, pos_filter = NULL, strict = TRUE, use_svydesign_weights = FALSE, use_svydesign_field = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE, exclude_nulls = FALSE, rename_nulls = "null_data", unique_colour = "indianred", title_size = 20, subtitle_size = 15 )fst_freq_compare( data, field, number = 10, norm = NULL, pos_filter = NULL, strict = TRUE, use_svydesign_weights = FALSE, use_svydesign_field = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE, exclude_nulls = FALSE, rename_nulls = "null_data", unique_colour = "indianred", title_size = 20, subtitle_size = 15 )
data |
A dataframe of text in CoNLL-U format with additional 'field' column for splitting data. |
field |
Column in 'data' used for splitting groups |
number |
The number of n-grams to return, default is '10'. |
norm |
The method for normalising the data. Valid settings are '"number_words"' (the number of words in the responses), '"number_resp"' (the number of responses), or 'NULL' (raw count returned, default, also used when weights are applied). |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' which means all word types included. |
strict |
Whether to strictly cut-off at 'number' (ties are alphabetically ordered), default is 'TRUE'. |
use_svydesign_weights |
Option to weight words in the wordcloud using weights from a svydesign object containing the raw data, default is 'FALSE' |
use_svydesign_field |
Option to get 'field' for splitting the data from the svydesign object, default is 'FALSE' |
id |
ID column from raw data, required if 'use_svydesign_weights = TRUE' and must match the 'docid' in formatted 'data'. |
svydesign |
A svydesign object which contains the raw data and weights. |
use_column_weights |
Option to weight words in the wordcloud using weights from formatted data which includes addition 'weight' column, default is 'FALSE' |
exclude_nulls |
Whether to include NULLs in 'field' column, default is 'FALSE' |
rename_nulls |
What to fill NULL values with if 'exclude_nulls = FALSE'. |
unique_colour |
Colour to display unique words, default is '"indianred"'. |
title_size |
size to display plot title |
subtitle_size |
size to display title of individual top words plot |
Plots of most frequent words in the plots pane with unique words highlighted.
fst_freq_compare(fst_child, 'gender', number = 10, norm = "number_resp") fst_freq_compare(fst_child, 'gender', number = 10, norm = NULL) s <- survey::svydesign(id=~1, weights= ~paino, data = child) c2 <- fst_child_2 c <- fst_child g <- 'gender' fst_freq_compare(c2, g, 10, NULL, NULL, TRUE, TRUE, TRUE, 'fsd_id', s) fst_freq_compare(c, g, use_column_weights = TRUE, strict = FALSE)fst_freq_compare(fst_child, 'gender', number = 10, norm = "number_resp") fst_freq_compare(fst_child, 'gender', number = 10, norm = NULL) s <- survey::svydesign(id=~1, weights= ~paino, data = child) c2 <- fst_child_2 c <- fst_child g <- 'gender' fst_freq_compare(c2, g, 10, NULL, NULL, TRUE, TRUE, TRUE, 'fsd_id', s) fst_freq_compare(c, g, use_column_weights = TRUE, strict = FALSE)
Plots most common words.
fst_freq_plot(table, number = NULL, name = NULL)fst_freq_plot(table, number = NULL, name = NULL)
table |
Output of 'fst_freq_table()' or 'fst_ngrams_table()'. |
number |
Optional number of n-grams for the title, default is 'NULL'. |
name |
An optional "name" for the plot to add to title, default is 'NULL'. |
Plot of top words.
pf <- c("NOUN", "VERB", "ADJ", "ADV") top_words <- fst_freq_table(fst_child, number = 15, pos_filter = pf) fst_freq_plot(top_words, number = 15, name = "Bullying")pf <- c("NOUN", "VERB", "ADJ", "ADV") top_words <- fst_freq_table(fst_child, number = 15, pos_filter = pf) fst_freq_plot(top_words, number = 15, name = "Bullying")
Creates a table of the most frequently-occurring words (unigrams) within the data. Optionally, weights can be provided either through a 'weight' column in the formatted data, or from a 'svydesign' object with the raw (preformatted) data.
fst_freq_table( data, number = 10, norm = NULL, pos_filter = NULL, strict = TRUE, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )fst_freq_table( data, number = 10, norm = NULL, pos_filter = NULL, strict = TRUE, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
number |
The number of top words to return, default is '10'. |
norm |
The method for normalising the data. Valid settings are '"number_words"' (the number of words in the responses), '"number_resp"' (the number of r , or 'NULL' (raw count returned, default, also used when weights are applied). |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' which means all word types included. |
strict |
Whether to strictly cut-off at 'number' (ties are alphabetically ordered), default is 'TRUE'. |
use_svydesign_weights |
Option to weight words in the table using weights from a 'svydesign' containing the raw data, default is 'FALSE' |
id |
ID column from raw data, required if 'use_svydesign_weights = TRUE' and must match the 'docid' in formatted 'data'. |
svydesign |
A 'svydesign' which contains the raw data and weights, required if 'use_svydesign_weights = TRUE'. |
use_column_weights |
Option to weight words in the table using weights from formatted data which includes addition 'weight' column, default is 'FALSE' |
A table of the most frequently occurring words in the data.
pf <- c("NOUN", "VERB", "ADJ", "ADV") pf2 <- "NOUN, VERB, ADJ, ADV" fst_freq_table(fst_child, number = 15, strict = FALSE, pos_filter = pf) fst_freq_table(fst_child, number = 15, strict = FALSE, pos_filter = pf2) fst_freq_table(fst_child, norm = 'number_words') fst_freq_table(fst_child, use_column_weights = TRUE) c2 <- fst_child_2 s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' fst_freq_table(c2, use_svydesign_weights = TRUE, svydesign = s, id = i)pf <- c("NOUN", "VERB", "ADJ", "ADV") pf2 <- "NOUN, VERB, ADJ, ADV" fst_freq_table(fst_child, number = 15, strict = FALSE, pos_filter = pf) fst_freq_table(fst_child, number = 15, strict = FALSE, pos_filter = pf2) fst_freq_table(fst_child, norm = 'number_words') fst_freq_table(fst_child, use_column_weights = TRUE) c2 <- fst_child_2 s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' fst_freq_table(c2, use_svydesign_weights = TRUE, svydesign = s, id = i)
Takes a list containing at least two tables of n-grams and frequencies (either output of 'fst_freq_table()' or 'fst_ngrams_table()') and finds n-grams unique to one table.
fst_get_unique_ngrams(list_of_top_ngrams)fst_get_unique_ngrams(list_of_top_ngrams)
list_of_top_ngrams |
A list of top ngrams |
Dataframe of words and whether word is unique or not.
top_child <- fst_freq_table(fst_child) top_dev <- fst_freq_table(fst_dev_coop) list_of_top_words <- list() list_of_top_words <- append(list_of_top_words, list(top_child)) list_of_top_words <- append(list_of_top_words, list(top_dev)) fst_get_unique_ngrams(list_of_top_words)top_child <- fst_freq_table(fst_child) top_dev <- fst_freq_table(fst_dev_coop) list_of_top_words <- list() list_of_top_words <- append(list_of_top_words, list(top_child)) list_of_top_words <- append(list_of_top_words, list(top_dev)) fst_get_unique_ngrams(list_of_top_words)
Takes at least two separate tables of n-grams and frequencies (either output of 'fst_freq_table()' or 'fst_ngrams_table()') and finds n-grams unique to one table.
fst_get_unique_ngrams_separate(table1, table2, ...)fst_get_unique_ngrams_separate(table1, table2, ...)
table1 |
The first n-grams table. |
table2 |
The second n-grams table. |
... |
Any other n-grams tables you want to include. |
Dataframe of words and whether word is unique or not.
top_child <- fst_freq_table(fst_child) top_dev <- fst_freq_table(fst_dev_coop) fst_get_unique_ngrams_separate(top_child, top_dev)top_child <- fst_freq_table(fst_child) top_dev <- fst_freq_table(fst_dev_coop) fst_get_unique_ngrams_separate(top_child, top_dev)
Merges list of unique words from 'fst_get_unique_ngrams()' with output of 'fst_freq_table()' or 'fst_ngrams_table()' so that unique words can be displayed on comparison plots.
fst_join_unique(table, unique_table)fst_join_unique(table, unique_table)
table |
Output of 'fst_freq_table()' or 'fst_ngrams_table()'. |
unique_table |
Output of 'fst_get_unique_ngrams()'. |
A table of top n-grams, frequency, and whether the n-gram is "unique".
top_child <- fst_freq_table(fst_child) top_dev <- fst_freq_table(fst_dev_coop) unique_words <- fst_get_unique_ngrams_separate(top_child, top_dev) fst_join_unique(top_child, unique_words) fst_join_unique(top_dev, unique_words)top_child <- fst_freq_table(fst_child) top_dev <- fst_freq_table(fst_dev_coop) unique_words <- fst_get_unique_ngrams_separate(top_child, top_dev) fst_join_unique(top_child, unique_words) fst_join_unique(top_dev, unique_words)
Compare length of text responses for different groups of participants. Data is split based on different values in the 'field' column of formatted data. Results will be shown within the plots pane.
fst_length_compare( data, field, incl_sentences = TRUE, exclude_nulls = FALSE, rename_nulls = "null_data" )fst_length_compare( data, field, incl_sentences = TRUE, exclude_nulls = FALSE, rename_nulls = "null_data" )
data |
A dataframe of text in CoNLL-U format with additional 'field' column for splitting data. |
field |
Column in 'data' used for splitting groups |
incl_sentences |
Whether to include sentence data in table, default is 'TRUE'. |
exclude_nulls |
Whether to include NULLs in 'field' column, default is 'FALSE' |
rename_nulls |
What to fill NULL values with if 'exclude_nulls = FALSE'. |
Dataframe summarising response lengths.
fst_length_compare(fst_child, 'gender') fst_length_compare(fst_dev_coop, 'education_level', incl_sentences = FALSE)fst_length_compare(fst_child, 'gender') fst_length_compare(fst_dev_coop, 'education_level', incl_sentences = FALSE)
Creates a table summarising distribution of the length of responses.
fst_length_summary(data, desc = "All responses", incl_sentences = TRUE)fst_length_summary(data, desc = "All responses", incl_sentences = TRUE)
data |
dataframe of text in CoNLL-U format, with optional additional columns. |
desc |
An optional string describing responses in table, default is '"All responses"'. |
incl_sentences |
Whether to include sentence data in table, default is 'TRUE'. |
Table summarising distribution of lengths of responses.
fst_length_summary(fst_child, incl_sentences = FALSE) fst_length_summary(fst_dev_coop, desc = "Q11_3")fst_length_summary(fst_child, incl_sentences = FALSE) fst_length_summary(fst_dev_coop, desc = "Q11_3")
Creates a plot of the most frequently-occurring n-grams within the data. Optionally, weights can be provided either through a 'weight' column in the formatted data, or from a 'svydesign' object with the raw (preformatted) data.
fst_ngrams( data, number = 10, ngrams = 1, norm = NULL, pos_filter = NULL, strict = TRUE, name = NULL, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )fst_ngrams( data, number = 10, ngrams = 1, norm = NULL, pos_filter = NULL, strict = TRUE, name = NULL, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
number |
The number of top words to return, default is '10'. |
ngrams |
The type of n-grams, default is '1'. |
norm |
The method for normalising the data. Valid settings are '"number_words"' (the number of words in the responses, default), '"number_resp"' (the number of responses), or 'NULL' (raw count returned). |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' which means all word types included. |
strict |
Whether to strictly cut-off at 'number' (ties are alphabetically ordered), default is 'TRUE'. |
name |
An optional "name" for the plot to add to title, default is 'NULL'. |
use_svydesign_weights |
Option to weight words in the plot using weights from a 'svydesign' containing the raw data, default is 'FALSE' |
id |
ID column from raw data, required if 'use_svydesign_weights = TRUE' and must match the 'docid' in formatted 'data'. |
svydesign |
A 'svydesign' which contains the raw data and weights, required if 'use_svydesign_weights = TRUE'. |
use_column_weights |
Option to weight words in the plot using weights from formatted data which includes addition 'weight' column, default is 'FALSE' |
Plot of top n-grams
fst_ngrams(fst_child, 12, ngrams = 2, strict = FALSE, name = "All") c <- fst_child_2 s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' T <- TRUE fst_ngrams(c, ngrams = 3, use_svydesign_weights = T, svydesign = s, id = i)fst_ngrams(fst_child, 12, ngrams = 2, strict = FALSE, name = "All") c <- fst_child_2 s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' T <- TRUE fst_ngrams(c, ngrams = 3, use_svydesign_weights = T, svydesign = s, id = i)
Find top and unique top n-grams for different groups of participants. Data is split based on different values in the 'field' column of formatted data. Results will be shown within the plots pane.
fst_ngrams_compare( data, field, number = 10, ngrams = 1, norm = NULL, pos_filter = NULL, strict = TRUE, use_svydesign_weights = FALSE, use_svydesign_field = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE, exclude_nulls = FALSE, rename_nulls = "null_data", unique_colour = "indianred", title_size = 20, subtitle_size = 15 )fst_ngrams_compare( data, field, number = 10, ngrams = 1, norm = NULL, pos_filter = NULL, strict = TRUE, use_svydesign_weights = FALSE, use_svydesign_field = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE, exclude_nulls = FALSE, rename_nulls = "null_data", unique_colour = "indianred", title_size = 20, subtitle_size = 15 )
data |
A dataframe of text in CoNLL-U format with additional 'field' column for splitting data. |
field |
Column in 'data' used for splitting groups |
number |
The number of n-grams to return, default is '10'. |
ngrams |
The type of n-grams to return, default is '1'. |
norm |
The method for normalising the data. Valid settings are '"number_words"' (the number of words in the responses), '"number_resp"' (the number of responses), or 'NULL' (raw count returned, default, also used when weights are applied). |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' which means all word types included. |
strict |
Whether to strictly cut-off at 'number' (ties are alphabetically ordered), default is 'TRUE'. |
use_svydesign_weights |
Option to weight words in the wordcloud using weights from a svydesign object containing the raw data, default is 'FALSE' |
use_svydesign_field |
Option to get 'field' for splitting the data from the svydesign object, default is 'FALSE' |
id |
ID column from raw data, required if 'use_svydesign_weights = TRUE' and must match the 'docid' in formatted 'data'. |
svydesign |
A svydesign object which contains the raw data and weights. |
use_column_weights |
Option to weight words in the wordcloud using weights from formatted data which includes addition 'weight' column, default is 'FALSE' |
exclude_nulls |
Whether to include NULLs in 'field' column, default is 'FALSE' |
rename_nulls |
What to fill NULL values with if 'exclude_nulls = FALSE'. |
unique_colour |
Colour to display unique words, default is '"indianred"'. |
title_size |
size to display plot title |
subtitle_size |
size to display title of individual top ngrams plot |
Plots of top n-grams in the plots pane with unique n-grams highlighted.
c <- fst_child g <- 'gender' fst_ngrams_compare(c, g, ngrams = 4, number = 10, norm = "number_resp") fst_ngrams_compare(c, g, ngrams = 2, number = 10, norm = NULL) s <- survey::svydesign(id=~1, weights= ~paino, data = child) c2 <- fst_child_2 fst_ngrams_compare(c2, g, 10, 3, NULL, NULL, TRUE, TRUE, TRUE, 'fsd_id', s) fst_ngrams_compare(c, g, 10, 2, use_column_weights = TRUE, strict = TRUE)c <- fst_child g <- 'gender' fst_ngrams_compare(c, g, ngrams = 4, number = 10, norm = "number_resp") fst_ngrams_compare(c, g, ngrams = 2, number = 10, norm = NULL) s <- survey::svydesign(id=~1, weights= ~paino, data = child) c2 <- fst_child_2 fst_ngrams_compare(c2, g, 10, 3, NULL, NULL, TRUE, TRUE, TRUE, 'fsd_id', s) fst_ngrams_compare(c, g, 10, 2, use_column_weights = TRUE, strict = TRUE)
Plots frequency n-grams with unique n-grams highlighted.
fst_ngrams_compare_plot( table, number = 10, ngrams = 1, unique_colour = "indianred", name = NULL, override_title = NULL, title_size = 20 )fst_ngrams_compare_plot( table, number = 10, ngrams = 1, unique_colour = "indianred", name = NULL, override_title = NULL, title_size = 20 )
table |
The table of n-grams, output of 'get_unique_ngrams()'. |
number |
The number of n-grams, default is '10'. |
ngrams |
The type of n-grams, default is '1'. |
unique_colour |
Colour to display unique words, default is '"indianred"'. |
name |
An optional "name" for the plot, default is 'NULL'. |
override_title |
An optional title to override the automatic one for the plot. Default is 'NULL'. If 'NULL', title of plot will be 'number' "Most Common 'Term'". 'Term' is "Words", "Bigrams", or "N-Grams" where N > 2. |
title_size |
size to display plot title |
Plot of top n-grams with unique terms highlighted.
top_child <- fst_freq_table(fst_child) top_dev <- fst_freq_table(fst_dev_coop) unique_words <- fst_get_unique_ngrams_separate(top_child, top_dev) top_child_u <- fst_join_unique(top_child, unique_words) top_dev_u <- fst_join_unique(top_dev, unique_words) fst_ngrams_compare_plot(top_child_u, ngrams = 1, name = "Child") fst_ngrams_compare_plot(top_dev_u, ngrams = 1, name = "Dev", title_size = 10)top_child <- fst_freq_table(fst_child) top_dev <- fst_freq_table(fst_dev_coop) unique_words <- fst_get_unique_ngrams_separate(top_child, top_dev) top_child_u <- fst_join_unique(top_child, unique_words) top_dev_u <- fst_join_unique(top_dev, unique_words) fst_ngrams_compare_plot(top_child_u, ngrams = 1, name = "Child") fst_ngrams_compare_plot(top_dev_u, ngrams = 1, name = "Dev", title_size = 10)
Plots frequency n-grams.
fst_ngrams_plot(table, number = NULL, ngrams = 1, name = NULL)fst_ngrams_plot(table, number = NULL, ngrams = 1, name = NULL)
table |
Output of 'fst_get_top_words()' or 'fst_get_top_ngrams()'. |
number |
Optional number of n-grams for title, default is 'NULL'. |
ngrams |
The type of n-grams, default is '1'. |
name |
An optional "name" for the plot to add to title, default is 'NULL'. |
Plot of top n-grams.
top_bigrams <- fst_ngrams_table(fst_child, ngrams = 2, number = 15) fst_ngrams_plot(top_bigrams, ngrams = 2, number = 15, name = "Children")top_bigrams <- fst_ngrams_table(fst_child, ngrams = 2, number = 15) fst_ngrams_plot(top_bigrams, ngrams = 2, number = 15, name = "Children")
Creates a table of the most frequently-occurring n-grams within the data. Optionally, weights can be provided either through a 'weight' column in the formatted data, or from a 'svydesign' object with the raw (preformatted) data.
fst_ngrams_table( data, number = 10, ngrams = 1, norm = NULL, pos_filter = NULL, strict = TRUE, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )fst_ngrams_table( data, number = 10, ngrams = 1, norm = NULL, pos_filter = NULL, strict = TRUE, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
number |
The number of n-grams to return, default is '10'. |
ngrams |
The type of n-grams to return, default is '1'. |
norm |
The method for normalising the data. Valid settings are '"number_words"' (the number of words in the responses), '"number_resp"' (the number of responses), or 'NULL' (raw count returned, default, also used when weights are applied). |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' which means all word types included. |
strict |
Whether to strictly cut-off at 'number' (ties are alphabetically ordered), default is 'TRUE'. |
use_svydesign_weights |
Option to weight words in the table using weights from a 'svydesign' containing the raw data, default is 'FALSE' |
id |
ID column from raw data, required if 'use_svydesign_weights = TRUE' and must match the 'docid' in formatted 'data'. |
svydesign |
A 'svydesign' which contains the raw data and weights, required if 'use_svydesign_weights = TRUE'. |
use_column_weights |
Option to weight words in the table using weights from formatted data which includes addition 'weight' column, default is 'FALSE' |
A table of the most frequently occurring n-grams in the data.
pf <- c("NOUN", "VERB", "ADJ", "ADV") pf2 <- "NOUN, VERB, ADJ, ADV" fst_ngrams_table(fst_child, norm = NULL) fst_ngrams_table(fst_child, ngrams = 2, norm = "number_resp") fst_ngrams_table(fst_child, ngrams = 2, pos_filter = pf) fst_ngrams_table(fst_child, ngrams = 2, pos_filter = pf2) c2 <- fst_child_2 s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' fst_ngrams_table(c2, use_svydesign_weights = TRUE, svydesign = s, id = i) fst_ngrams_table(fst_child, use_column_weights = TRUE, ngrams = 3)pf <- c("NOUN", "VERB", "ADJ", "ADV") pf2 <- "NOUN, VERB, ADJ, ADV" fst_ngrams_table(fst_child, norm = NULL) fst_ngrams_table(fst_child, ngrams = 2, norm = "number_resp") fst_ngrams_table(fst_child, ngrams = 2, pos_filter = pf) fst_ngrams_table(fst_child, ngrams = 2, pos_filter = pf2) c2 <- fst_child_2 s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' fst_ngrams_table(c2, use_svydesign_weights = TRUE, svydesign = s, id = i) fst_ngrams_table(fst_child, use_column_weights = TRUE, ngrams = 3)
Creates a table of the most frequently-occurring n-grams within the data. Optionally, weights can be provided either through a 'weight' column in the formatted data, or from a 'svydesign' object with the raw (preformatted) data. Equivalent to ‘fst_get_top_ngrams' but doesn’t print message about ties.
fst_ngrams_table2( data, number = 10, ngrams = 1, norm = NULL, pos_filter = NULL, strict = TRUE, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )fst_ngrams_table2( data, number = 10, ngrams = 1, norm = NULL, pos_filter = NULL, strict = TRUE, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
number |
The number of n-grams to return, default is '10'. |
ngrams |
The type of n-grams to return, default is '1'. |
norm |
The method for normalising the data. Valid settings are '"number_words"' (the number of words in the responses, default), '"number_resp"' (the number of responses), or 'NULL' (raw count returned). |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' which means all word types included. |
strict |
Whether to strictly cut-off at 'number' (ties are alphabetically ordered), default is 'TRUE'. |
use_svydesign_weights |
Option to weight words in the table using weights from a 'svydesign' containing the raw data, default is 'FALSE' |
id |
ID column from raw data, required if 'use_svydesign_weights = TRUE' and must match the 'docid' in formatted 'data'. |
svydesign |
A 'svydesign' which contains the raw data and weights, required if 'use_svydesign_weights = TRUE'. |
use_column_weights |
Option to weight words in the table using weights from formatted data which includes addition 'weight' column, default is 'FALSE' |
A table of the most frequently occurring n-grams in the data.
fst_ngrams_table2(fst_child, norm = NULL) fst_ngrams_table2(fst_child, ngrams = 2, norm = "number_resp") c <- fst_child_2 s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' T <- TRUE fst_ngrams_table2(c, 10, 2, use_svydesign_weights = T, svydesign = s, id = i)fst_ngrams_table2(fst_child, norm = NULL) fst_ngrams_table2(fst_child, ngrams = 2, norm = "number_resp") c <- fst_child_2 s <- survey::svydesign(id=~1, weights= ~paino, data = child) i <- 'fsd_id' T <- TRUE fst_ngrams_table2(c, 10, 2, use_svydesign_weights = T, svydesign = s, id = i)
Creates a summary table for the input CoNLL-U data which counts the number of words of each part-of-speech tag within the data.
fst_pos(data)fst_pos(data)
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
A dataframe with a count and proportion of each UPOS tag in the data and the full name of the tag.
fst_pos(fst_child) fst_pos(fst_dev_coop)fst_pos(fst_child) fst_pos(fst_dev_coop)
Count each POS type for different groups of participants. Data is split based on different values in the 'field' column of formatted data. Results will be shown within the plots pane.
fst_pos_compare(data, field, exclude_nulls = FALSE, rename_nulls = "null_data")fst_pos_compare(data, field, exclude_nulls = FALSE, rename_nulls = "null_data")
data |
A dataframe of text in CoNLL-U format with additional 'field' column for splitting data. |
field |
Column in 'data' used for splitting groups |
exclude_nulls |
Whether to include NULLs in 'field' column, default is 'FALSE' |
rename_nulls |
What to fill NULL values with if 'exclude_nulls = FALSE'. |
Table of POS tag counts for the groups.
fst_pos_compare(fst_child, 'gender') fst_pos_compare(fst_dev_coop, 'region')fst_pos_compare(fst_child, 'gender') fst_pos_compare(fst_dev_coop, 'region')
Creates a dataframe in CoNLL-U format from a dataframe containing text from using the [udpipe] package and a language model plus any additional columns that are included such as 'weights' or columns added through 'add_cols'. Stopwords and punctuation are optionally removed if the the 'stopword_list' argument is not "none".
fst_prepare( data, question, id, model = "ftb", stopword_list = "nltk", language = "fi", weights = NULL, add_cols = NULL, manual = FALSE, manual_list = "" )fst_prepare( data, question, id, model = "ftb", stopword_list = "nltk", language = "fi", weights = NULL, add_cols = NULL, manual = FALSE, manual_list = "" )
data |
A dataframe of survey responses which contains an open-ended question. |
question |
The column in the dataframe which contains the open-ended question. |
id |
The column in the dataframe which contains the ids for the responses. |
model |
A language model available for [udpipe]. '"ftb"' (default) or '"tdt"' are recognised as shorthand for "finnish-ftb" and "finnish-tdt". The full list is available in the [udpipe] documentation or via 'fst_print_available_models()'. |
stopword_list |
A valid stopword list, default is '"nltk"', '"manual"' can be used to indicate that a manual list will be provided, or ‘"none"' if you don’t want to remove stopwords known as 'source' in 'stopwords::stopwords' |
language |
two-letter ISO code for the language for the stopword list |
weights |
Optional, the column of the dataframe which contains the respective weights for each response. |
add_cols |
Optional, a column (or columns) from the dataframe which contain other information you'd like to retain (for instance, dimension columnns for splitting the data for comparison plots). |
manual |
An optional boolean to indicate that a manual list will be provided, 'stopword_list = "manual"' can also or instead be used. |
manual_list |
A manual list of stopwords. |
'fst_prepare_conllu()' produces a dataframe containing survey text responses in CoNLL-U format with stopwords optionally removed.
A dataframe of text in CoNLL-U format.
## Not run: i <- "fsd_id" cb <- child dev <- dev_coop fst_prepare(data = cb, question = "q7", id = 'fsd_id', weights = 'paino') fst_prepare(data = dev, question = "q11_2", id = i, add_cols = c('gender')) fst_prepare(data = dev, question = "q11_3", id = i, add_cols = 'gender') fst_prepare(data = child, question = "q7", id = i, model = 'swedish-lines') unlink("finnish-ftb-ud-2.5-191206.udpipe") unlink("finnish-tdt-ud-2.5-191206.udpipe") unlink("swedish-lines-ud-2.5-191206.udpipe") ## End(Not run)## Not run: i <- "fsd_id" cb <- child dev <- dev_coop fst_prepare(data = cb, question = "q7", id = 'fsd_id', weights = 'paino') fst_prepare(data = dev, question = "q11_2", id = i, add_cols = c('gender')) fst_prepare(data = dev, question = "q11_3", id = i, add_cols = 'gender') fst_prepare(data = child, question = "q7", id = i, model = 'swedish-lines') unlink("finnish-ftb-ud-2.5-191206.udpipe") unlink("finnish-tdt-ud-2.5-191206.udpipe") unlink("swedish-lines-ud-2.5-191206.udpipe") ## End(Not run)
Creates a dataframe in CoNLL-U format from a 'svydesign' object including text using the [udpipe] package and a language model plus weights if these are included in the 'svydesign' object and any columns added through 'add_cols'.Stopwords and punctuation are optionally removed if the the 'stopword_list' argument is not "none".
fst_prepare_svydesign( svydesign, question, id, model = "ftb", stopword_list = "nltk", language = "fi", use_weights = TRUE, add_cols = NULL, manual = FALSE, manual_list = "" )fst_prepare_svydesign( svydesign, question, id, model = "ftb", stopword_list = "nltk", language = "fi", use_weights = TRUE, add_cols = NULL, manual = FALSE, manual_list = "" )
svydesign |
A 'svydesign' object which contains an open-ended question. |
question |
The column in the dataframe which contains the open-ended question. |
id |
The column in the dataframe which contains the ids for the responses. |
model |
A language model available for [udpipe]. '"ftb"' (default) or '"tdt"' are recognised as shorthand for "finnish-ftb" and "finnish-tdt". The full list is available in the [udpipe] documentation or via 'fst_print_available_models()'. |
stopword_list |
A valid stopword list, default is '"nltk"', or '"none"'. |
language |
two-letter ISO code for the language for the stopword list |
use_weights |
Optional, whether to use weights within the 'svydesign' |
add_cols |
Optional, a column (or columns) from the dataframe which contain other information you'd like to retain (for instance, dimension columnns for splitting the data for comparison plots). |
manual |
An optional boolean to indicate that a manual list will be provided, 'stopword_list = "manual"' can also or instead be used. |
manual_list |
A manual list of stopwords. |
'fst_prepare_svydesign()' produces a dataframe containing survey text responses in CoNLL-U format with stopwords optionally removed.
A dataframe of text in CoNLL-U format.
## Not run: i <- "fsd_id" svy_child <- survey::svydesign(id=~1, weights= ~paino, data = child) fst_prepare_svydesign(svy_child, question = "q7", id = i, use_weights = TRUE) svy_d <- survey::svydesign(id = ~1, weights = ~paino, data =dev_coop) fst_prepare_svydesign(svy_d, question = "q11_2", id = i, add_cols = 'gender') fst_prepare_svydesign(svy_d, 'q11_2', i, 'finnish-ftb', 'nltk', 'fi') unlink("finnish-ftb-ud-2.5-191206.udpipe") unlink("finnish-tdt-ud-2.5-191206.udpipe") ## End(Not run)## Not run: i <- "fsd_id" svy_child <- survey::svydesign(id=~1, weights= ~paino, data = child) fst_prepare_svydesign(svy_child, question = "q7", id = i, use_weights = TRUE) svy_d <- survey::svydesign(id = ~1, weights = ~paino, data =dev_coop) fst_prepare_svydesign(svy_d, question = "q11_2", id = i, add_cols = 'gender') fst_prepare_svydesign(svy_d, 'q11_2', i, 'finnish-ftb', 'nltk', 'fi') unlink("finnish-ftb-ud-2.5-191206.udpipe") unlink("finnish-tdt-ud-2.5-191206.udpipe") ## End(Not run)
Find treebanks available for use
fst_print_available_models(search = NULL)fst_print_available_models(search = NULL)
search |
An optional string for filtering the list, name of language in English, eg. 'estonian' |
List of available treebanks, filtered
fst_print_available_models() fst_print_available_models(search = "swedish")fst_print_available_models() fst_print_available_models(search = "swedish")
Removes stopwords and punctuation from a dataframe containing survey text data which is already in CoNLL-U format.
fst_rm_stop_punct( data, stopword_list = "nltk", language = "fi", manual = FALSE, manual_list = "" )fst_rm_stop_punct( data, stopword_list = "nltk", language = "fi", manual = FALSE, manual_list = "" )
data |
A dataframe of text in CoNLL-U format. |
stopword_list |
A valid stopword list, default is '"nltk"', '"manual"' can be used to indicate that a manual list will be provided, or ‘"none"' if you don’t want to remove stopwords, known as 'source' in 'stopwords::stopwords' |
language |
two-letter ISO code of the language for the stopword list |
manual |
An optional boolean to indicate that a manual list will be provided, 'stopword_list = "manual"' can also or instead be used. |
manual_list |
A manual list of stopwords. |
A dataframe of text in CoNLL-U format without stopwords and punctuation.
## Not run: c <- fst_format(child, question = 'q7', id = 'fsd_id') fst_rm_stop_punct(c) fst_rm_stop_punct(c, stopword_list = "snowball") fst_rm_stop_punct(c, "stopwords-iso") mlist <- c('en', 'et', 'ei', 'emme', 'ette', 'eivät', 'minä', 'minum') mlist2 <- "en, et, ei, emme, ette, eivät, minä, minum" fst_rm_stop_punct(c, manual = TRUE, manual_list = mlist) fst_rm_stop_punct(c, stopword_list = "manual", manual_list = mlist) unlink("finnish-ftb-ud-2.5-191206.udpipe") ## End(Not run)## Not run: c <- fst_format(child, question = 'q7', id = 'fsd_id') fst_rm_stop_punct(c) fst_rm_stop_punct(c, stopword_list = "snowball") fst_rm_stop_punct(c, "stopwords-iso") mlist <- c('en', 'et', 'ei', 'emme', 'ette', 'eivät', 'minä', 'minum') mlist2 <- "en, et, ei, emme, ette, eivät, minä, minum" fst_rm_stop_punct(c, manual = TRUE, manual_list = mlist) fst_rm_stop_punct(c, stopword_list = "manual", manual_list = mlist) unlink("finnish-ftb-ud-2.5-191206.udpipe") ## End(Not run)
Creates a summary table for the input CoNLL-U data which provides the response count and proportion, total number of words, the number of unique words, and the number of unique lemmas.
fst_summarise(data, desc = "All responses")fst_summarise(data, desc = "All responses")
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
desc |
A string describing responses in table, default is '"All responses"'. |
A dataframe with summary information for the data including response rate and word counts.
fst_summarise(fst_child) fst_summarise(fst_dev_coop, "Q11_3")fst_summarise(fst_child) fst_summarise(fst_dev_coop, "Q11_3")
Compare text responses for different groups of participants. Data is split based on different values in the 'field' column of formatted data. Results will be shown within the plots pane.
fst_summarise_compare( data, field, exclude_nulls = FALSE, rename_nulls = "null_data" )fst_summarise_compare( data, field, exclude_nulls = FALSE, rename_nulls = "null_data" )
data |
A dataframe of text in CoNLL-U format with additional 'field' column for splitting data. |
field |
Column in 'data' used for splitting groups |
exclude_nulls |
Whether to include NULLs in 'field' column, default is 'FALSE' |
rename_nulls |
What to fill NULL values with if 'exclude_nulls = FALSE'. |
Summary table of responses between groups.
fst_summarise_compare(fst_child, 'gender') fst_summarise_compare(fst_dev_coop, 'gender')fst_summarise_compare(fst_child, 'gender') fst_summarise_compare(fst_dev_coop, 'gender')
Creates a summary table for the input CoNLL-U data which provides the total number of words, the number of unique words, and the number of unique lemmas.
fst_summarise_short(data)fst_summarise_short(data)
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
A dataframe with summary information on word counts for the data.
fst_summarise_short(fst_child) fst_summarise_short(fst_dev_coop)fst_summarise_short(fst_child) fst_summarise_short(fst_dev_coop)
This function takes data in CoNLL-U format and a 'svydesign' (from 'survey' package) object with weights in it and merges the weights, and any additional columns into the formatted data.
fst_use_svydesign(data, svydesign, id, add_cols = NULL, add_weights = TRUE)fst_use_svydesign(data, svydesign, id, add_cols = NULL, add_weights = TRUE)
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
svydesign |
A 'svydesign' object containing the raw data which produced the 'data' |
id |
ID column from raw data, must match the 'docid' in formatted 'data' |
add_cols |
Optional, a column (or columns) from the dataframe which contain other information you'd need (for instance, covariate column for splitting the data for comparison plots). |
add_weights |
Optional, a boolean for whether to add weights from svydesign object, default is 'TRUE'. |
A dataframe of text in CoNLL-U format plus a ''weight'' column and optional other columns
svy_child <- survey::svydesign(id=~1, weights= ~paino, data = child) fst_use_svydesign(data = fst_child_2, svydesign = svy_child, id = 'fsd_id') svy_dev <- survey::svydesign(id = ~1, weights = ~paino, data = dev_coop) fst_use_svydesign(data = fst_dev_coop_2, svydesign = svy_dev, id = 'fsd_id')svy_child <- survey::svydesign(id=~1, weights= ~paino, data = child) fst_use_svydesign(data = fst_child_2, svydesign = svy_child, id = 'fsd_id') svy_dev <- survey::svydesign(id = ~1, weights = ~paino, data = dev_coop) fst_use_svydesign(data = fst_dev_coop_2, svydesign = svy_dev, id = 'fsd_id')
Creates a wordcloud from CoNLL-U data of frequently-occurring words. Optionally, weights can be provided either through a 'weight' column in the formatted data, or from a 'svydesign' object with the raw (preformatted) data.
fst_wordcloud( data, pos_filter = NULL, max = 100, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )fst_wordcloud( data, pos_filter = NULL, max = 100, use_svydesign_weights = FALSE, id = "", svydesign = NULL, use_column_weights = FALSE )
data |
A dataframe of text in CoNLL-U format, with optional additional columns. |
pos_filter |
List of UPOS tags for inclusion, default is 'NULL' which means all word types included. |
max |
The maximum number of words to display, default is '100'. |
use_svydesign_weights |
Option to weight words in the wordcloud using weights from a 'svydesign' containing the raw data, default is 'FALSE' |
id |
ID column from raw data, required if 'use_svydesign_weights = TRUE' and must match the 'docid' in formatted 'data'. |
svydesign |
A 'svydesign' which contains the raw data and weights, required if 'use_svydesign_weights = TRUE'. |
use_column_weights |
Option to weight words in the wordcloud using weights from formatted data which includes addition 'weight' column, default is 'FALSE'. |
A wordcloud from the data.
fst_wordcloud(fst_child) fst_wordcloud(fst_child, pos_filter = c("NOUN", "VERB", "ADJ", "ADV")) fst_wordcloud(fst_child, pos_filter = 'NOUN, VERB, ADJ') fst_wordcloud(fst_child, use_column_weights = TRUE) i <- 'fsd_id' c <- fst_child_2 s <- survey::svydesign(id=~1, weights= ~paino, data = child) fst_wordcloud(c, use_svydesign_weights = TRUE, id = i, svydesign = s)fst_wordcloud(fst_child) fst_wordcloud(fst_child, pos_filter = c("NOUN", "VERB", "ADJ", "ADV")) fst_wordcloud(fst_child, pos_filter = 'NOUN, VERB, ADJ') fst_wordcloud(fst_child, use_column_weights = TRUE) i <- 'fsd_id' c <- fst_child_2 s <- survey::svydesign(id=~1, weights= ~paino, data = child) fst_wordcloud(c, use_svydesign_weights = TRUE, id = i, svydesign = s)
Run Shiny App Demo
runDemo()runDemo()
launches the RShiny demo
## Not run: runDemo() ## End(Not run)## Not run: runDemo() ## End(Not run)