Skip to content

Releases: michael-rapp/Apriori

Version 2.0.1

12 Apr 22:22
Compare
Choose a tag to compare

A minor release that introduces the following changes:

  • Updated Kotlin to version 1.3.21.
  • Updated Gradle configuration.

Version 2.0.0

30 Aug 21:31
Compare
Choose a tag to compare

A major release, which introduces the following changes:

  • Migrated the library to use the Kotlin programming language instead of Java.

Version 1.3.0

13 Dec 22:35
Compare
Choose a tag to compare

A feature release, which introduces the following changes:

  • The execute-method of the class Apriori now takes an Iterable as an argument. This allows to properly iterate transactions multiple times.
  • In accordance with the API specification, iterators are now expected to throw a NoSuchElementException instead of returning null when the end is reached.

Version 1.2.0

08 Aug 22:57
Compare
Choose a tag to compare

A feature release, which introduces the following changes:

  • An enhanced API for sorting and filtering association rules and frequent item sets has been added
  • The class AssociationRule does now provide covers-methods, which allow to test, if the rule applies to a set of items
  • Added support for the conviction metric

Version 1.1.1

31 Jul 20:44
Compare
Choose a tag to compare

A bugfix release, which fixes the following issues:

Version 1.1.0

30 Jul 01:24
Compare
Choose a tag to compare

A feature release, which introduces the following changes:

  • Added the ability to use different tie-breaking strategies when sorting or filtering rule sets

Version 1.0.1

28 Jul 18:42
Compare
Choose a tag to compare

A bugfix release, which fixed the following issues:

Version 1.0.0

16 Jun 21:25
Compare
Choose a tag to compare

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

  • Allows to find frequent item sets and (optionally) to generate association rules using the Apriori algorithm
  • The Apriori algorithm can be configured using the builder pattern
  • The algorithm can be configured to try to find a specific number of frequent item sets and/or association rules
  • Rule sets can be ordered and filtered by support, confidence, lift and leverage
  • Logging is provided by using the SLF4J logging facade