Version 1.1.3

NEWS

Versioning

Releases will be numbered with the following semantic versioning format:

<major>.<minor>.<patch>

And constructed with the following guidelines:

• Breaking backward compatibility bumps the major (and resets the minor
  and patch)
• New additions without breaking backward compatibility bumps the minor
  (and resets the patch)
• Bug fixes and misc changes bumps the patch

lexicon 1.0.1 - 1.1.3

BUG FIXES

• hash_lemmas had the lemma of as to be a. This was incorrect (spotted by
  Jonathan Bratt).

• hash_lemmas had Spaces before 2 tokens (" furtherst", " skilled") meaning.
  This extra white space has been stripped.

• The hash_sentiment_senticnett dictionary contained "sparsely" which is also
  contained in hash_valence_shifters. This term has been dropped from the
  hash_sentiment_senticnett dictionary. See # 12 for more info.

NEW FEATURES

• profanity_zac_anger added to provide a longer list of profane words.

• profanity_racist added to provide a profane list that is specific for
  detecting racist terms.

• key_regressive_imagery added to provide R users with access to Colin
  Martindale's (1975, 1990) English Regressive Imagery Dictionary (RID). The
  Regressive Imagery Dictionary (RID) is a text analysis coding taxonomy that
  can be used to measure the degree to which a text is primordial vs.
  conceptual.

• key_corporate_social_responsibility added to provide R users with access to
  Pencle & Mălăescu's Corporate Social Responsibility (CSR) Dictionary.

MINOR FEATURES

• available_data picks up a regex argument to search for specific substrings
  and return matching rows.

IMPROVEMENTS

• hash_sentiment_jockers_rinker now contains the word 'fuckin'. Additionally,
  the word 'fucking' has a milder negative value because this word, though often
  used as a negator, is also used as a amplifier. By reducing it's weight it
  allows more positive words to have more pull but if no polarized words exist
  'fucking' will still keep the typical negative direction of the clause.

Version 0.7.0

NEWS

Versioning

Releases will be numbered with the following semantic versioning format:

<major>.<minor>.<patch>

And constructed with the following guidelines:

• Breaking backward compatibility bumps the major (and resets the minor
  and patch)
• New additions without breaking backward compatibility bumps the minor
  (and resets the patch)
• Bug fixes and misc changes bumps the patch

lexicon 0.7.0 -

lexicon 0.5.0 - 0.6.3

NEW FEATURES

• sw_mallete, sw_jockers, sw_python, sw_lucene, sw_loughran_mcdonald_short,
  & sw_loughran_mcdonald_long stopword lists added.

• hash_sentiment_senticnet hash_sentiment_vadar, hash_sentiment_inquirer,
  hash_sentiment_loughran_mcdonald, hash_sentiment_emojis &
  hash_sentiment_jockers_rinker sentiment hash tables added for use in the
  sentimentr package.

• modal_loughran_mcdonald added; a data.table of weak, moderate, and strong
  modal verbs.

• constraining_loughran_mcdonald added, a vector of words that are associated
  with constraining.

• hash_emojis and emojis_sentiment data sets added for text analysis with
  emojis.

IMPROVEMENTS

• hash_valence_shifters added following negators: "daren't", "hadn't",
  "needn't", "oughtn't"; the following amplifiers: "absolutely", "considerably",
  "decidedly", "especially", "majorly", "most", "uber"; the following
  de-amplifiers: "almost", "kind of", "kinda", "partly", "somewhat", "sort of",
  "sorta". In addition, all contraction negators were re-added
  to the hash_valence_shifters sans apostrophe as cleaning or less formal
  writing may result in contractions without apostrophes.

CHANGES

• The word "incredibly" was removed from hash_sentiment_jockers,
  hash_sentiment_huliu, & hash_sentiment_sentiword and added as an amplifier
  to hash_valence_shifters. Spotted by AbdulMajedRaja:
  trinker/sentimentr#58 (comment)

lexicon 0.4.0 - 0.4.1

BUG FIXES

• function_words contained duplicates that have been been removed.

• hash_lemmas contained an erroneous token-lemma pair (also-conjurer). This
  was spotted by Mitchell Linegar (see trinker/textstem#5).
  The token also has been removed from the dictionary.

NEW FEATURES

• pos_df_irregular_nouns and pos_unchanging_nouns added. The former is a
  data.frame of singular and plural forms of irregular nouns. The latter is
  a simple list of irregular nouns that have the same singular and plural forms.

• profanity_alvarez, profanity_arr_bad, profanity_banned,
  profanity_google, & profanity_von_ahn added to give access to profanity
  word lists.

lexicon 0.3.0 - 0.3.1

BUG FIXES

• freq_first_names and freq_last_names were just a string of the data set
  name. This has been updated with the actual data set.

NEW FEATURES

• available_data added to see what data sets are available in lexicon.

lexicon 0.2.0

NEW FEATURES

• hash_sentiment_jockers and key_sentiment_jockers added as objects though
  they are not data objects but for all purposes act the same. These data sets
  come from syuzhet's custom dictionary built by Jockers.

CHANGES

• hash_sentiment and hash_sentiword renamed to hash_sentiment_huliu and
  hash_sentiment_sentiword for consistency.

lexicon 0.1.1

NEW FEATURES

• hash_grady_pos added to provide a lookup of Grady's parts of speech for words.

• hash_lemmas added to provide a lookup of Mechura's lemmatization list.

• hash_sentiment_jockers and key_sentiment_jockers added as objects though
  they are not data objects but for all purposes act the same. These data sets
  come from syuzhet's custom dictionary built by Jockers.

lexicon 0.1.0

NEW FEATURES

• The ratings and grades keys from sentimentr have been moved to the
  lexicon package and renamed to key_rating and key_grade.

IMPROVEMENTS

• Added the positve terms 'spot on', 'on time', & 'on point' to hash_sentiment.

lexicon 0.0.1

This package is a collection of lexical hash tables, dictionaries, and word
lists.

version 0.3.1

NEWS

Versioning

Releases will be numbered with the following semantic versioning format:

<major>.<minor>.<patch>

And constructed with the following guidelines:

• Breaking backward compatibility bumps the major (and resets the minor
  and patch)
• New additions without breaking backward compatibility bumps the minor
  (and resets the patch)
• Bug fixes and misc changes bumps the patch

lexicon 0.3.0 - 0.3.1

BUG FIXES

• freq_first_names and freq_last_names were just a string of the data set
  name. This has been updated with the actual data set.

NEW FEATURES

• available_data added to see what data sets are available in lexicon.

lexicon 0.2.0

NEW FEATURES

• hash_sentiment_jockers and key_sentiment_jockers added as objects though
  they are not data objects but for all purposes act the same. These data sets
  come from syuzhet's custom dictionary built by Jockers.

CHANGES

• hash_sentiment and hash_sentiword renamed to hash_sentiment_huliu and
  hash_sentiment_sentiword for consistency.

lexicon 0.1.1

NEW FEATURES

• hash_grady_pos added to provide a lookup of Grady's parts of speech for words.

• hash_lemmas added to provide a lookup of Mechura's lemmatization list.

• hash_sentiment_jockers and key_sentiment_jockers added as objects though
  they are not data objects but for all purposes act the same. These data sets
  come from syuzhet's custom dictionary built by Jockers.

lexicon 0.1.0

NEW FEATURES

• The ratings and grades keys from sentimentr have been moved to the
  lexicon package and renamed to key_rating and key_grade.

IMPROVEMENTS

• Added the positve terms 'spot on', 'on time', & 'on point' to hash_sentiment.

lexicon 0.0.1

This package is a collection of lexical hash tables, dictionaries, and word
lists.

lexicon version 0.2.0

NEWS

Versioning

Releases will be numbered with the following semantic versioning format:

<major>.<minor>.<patch>

And constructed with the following guidelines:

• Breaking backward compatibility bumps the major (and resets the minor
  and patch)
• New additions without breaking backward compatibility bumps the minor
  (and resets the patch)
• Bug fixes and misc changes bumps the patch

lexicon 0.2.0

NEW FEATURES

• hash_sentiment_jockers and key_sentiment_jockers added as objects though
  they are not data objects but for all purposes act the same. These data sets
  come from syuzhet's custom dictionary built by Jockers.

CHANGES

• hash_sentiment and hash_sentiword renamed to hash_sentiment_huliu and
  hash_sentiment_sentiword for consistency.

lexicon 0.1.1

NEW FEATURES

• hash_grady_pos added to provide a lookup of Grady's parts of speech for words.

• hash_lemmas added to provide a lookup of Mechura's lemmatization list.

• hash_sentiment_jockers and key_sentiment_jockers added as objects though
  they are not data objects but for all purposes act the same. These data sets
  come from syuzhet's custom dictionary built by Jockers.

lexicon 0.1.0

NEW FEATURES

• The ratings and grades keys from sentimentr have been moved to the
  lexicon package and renamed to key_rating and key_grade.

IMPROVEMENTS

• Added the positve terms 'spot on', 'on time', & 'on point' to hash_sentiment.

lexicon 0.0.1

This package is a collection of lexical hash tables, dictionaries, and word
lists.

lexicon 0.1.1

NEWS

Versioning

Releases will be numbered with the following semantic versioning format:

<major>.<minor>.<patch>

And constructed with the following guidelines:

• Breaking backward compatibility bumps the major (and resets the minor
  and patch)
• New additions without breaking backward compatibility bumps the minor
  (and resets the patch)
• Bug fixes and misc changes bumps the patch

lexicon 0.1.1

NEW FEATURES

• hash_grady_pos added to provide a lookup of Grady's parts of speech for words.
• hash_lemmas added to provide a lookup of Mechura's lemmatization list.

lexicon 0.1.0

NEW FEATURES

• The ratings and grades keys from sentimentr have been moved to the
  lexicon package and renamed to key_rating and key_grade.

IMPROVEMENTS

• Added the positve terms 'spot on', 'on time', & 'on point' to hash_sentiment.

lexicon 0.0.1

This package is a collection of lexical hash tables, dictionaries, and word
lists.