Skip to content

Latest commit

 

History

History
76 lines (49 loc) · 3.35 KB

RELEASE_NOTES.md

File metadata and controls

76 lines (49 loc) · 3.35 KB
Raw

TextMiningUtil - RELEASE NOTES

Version 2.1.3 (Oct. 23th 2019)

A minor release, which introduces the following changes:

  • Added positional-methods to interface Processor.
  • Updated dependency "JavaUtil" to version 2.4.0.
  • Updated Kotlin to version 1.3.50.

Version 2.1.2 (Feb. 23th 2019)

A minor release, which introduces the following changes:

  • Updated dependency "JavaUtil" to version 2.1.1.
  • Updated Kotlin to version 1.3.21.

Version 2.1.1 (Jan. 28th 2019)

A minor release, which introduces the following changes:

  • Tokenizers do now handle empty tokens properly.
  • Add predefined Processor for removing leading and trailing whitespace from tokens.

Version 2.1.0 (Jan. 27th 2019)

A feature release, which introduces the following changes:

  • Added interface TextParser as well as the implementations AbstractTextParser and GradualTextParser.
  • Added types of tokens, such as MutableToken, ValueToken and TokenSequence, that are useful when parsing texts.
  • Added class Dictionary that allows to translate parts of texts using key-value pairs.
  • Added class DictionaryTokenizer that allows to split texts into tokens by using a Dictionary.
  • Added interface Processor and class ProcessorChain.
  • Added a helper method to create case-insensitive metrics to the interface TextMetric.
  • Added class TextMetric.Comparator.
  • Tokenizers and TextMetrics can now be applied to CharSequences.
  • Replaced getter and setter methods with properties to be in accordance with the Kotlin paradigm
  • Updated dependency "JavaUtil" to version 2.0.2.
  • Updated Kotlin to version 1.3.11.

Version 2.0.0 (Aug. 7th 2018)

A major release, which introduces the following changes:

  • Migrated the project to use the Kotlin programming language instead of Java.
  • Converted the inner interface Tokenizer.Token into a separate interface Token.
  • Converted the inner class NGramTokenizer.NGram into a separate class NGram.
  • Changed the return type of the method Tokenizer#tokenize from Set to Collection.

Version 1.2.0 (May. 21th 2018)

A feature release, which introduces the following changes:

  • Added the metrics OptimalStringAlignmentDistance, OptimalStringAlignmentDissimilarity and OptimalStringAlignmentSimilarity.
  • Added the metrics DamerauLevenshteinDistance, DamerauLevenshteinDissimilarity and DamerauLevenshteinSimilarity.
  • Added the tokenizers FixedLengthTokenizer and RegexTokenizer.

Version 1.1.1 (Apr. 29th 2018)

A minor release, which provides the following changes:

  • Updated dependency "JavaUtil" to version 1.2.0.

Version 1.1.0 (Nov. 19th 2017)

A feature release, which provides the following changes:

  • Tokenizers do not return multiple n-grams or substrings with the same token anymore. Instead, the positions off all duplicates are aggregated in one NGram or Substring.
  • Added constructors, which only allows to specify a maximum length, but no minimum length, to the classes NGramTokenizer and DiceCoefficient.

Version 1.0.0 (Nov. 17th 2017)

The first stable release of the library, which provides the following utility classes:

  • The classes SubstringTokenizer and NGramTokenizer for splitting texts into shorter subtexts.
  • The metrics DiceCoefficient, HammingDistance, HammingLoss, HammingAccuracy, LevenshteinDistance, LevenshteinDissimilarity and LevenshteinSimilarity.