The Intelligent Playlist plugin functions by dropping just some tracks into an initial playlist.
A short warning beforehand: I have written this script for an Artificial Intelligence lecture at my university. Thus, its main goal is to apply some typical algorithms and ideas from AI in a real-life software project. If you are seeking for something to integrate typical "Similar Artists" functionality provided by online services, you are better off with Amarok's Dynamic Playlists (and some custom biases) or something like this:
The Intelligent Playlist plugin functions by dropping just some tracks into an initial playlist. Turn the plugin on and hit the play button. The plugin will then try to find similar tracks from your local collection. So, beneath of using a similarity comparison instead of a given (a-priory) set of rules ("biases") it shall not distinguish too much in how it presents itself to the user compared to Amarok's Dynamic Playlists.
For a detailed description, please have a look at the projects homepage.
Please note, that this is currently rarely tested. This means: It might work for your music flavor and collection or not. I have made pretty good experience, but I cannot guarantee that this will apply to everyone else. So, feedback and further suggestions about how to improve the plugin are very welcome.
· KDE Software Compilation
What's New in This Release: [ read full changelog ]
· plugin is now localizable
· added Czech translations (thanks to Pavel Fric)
· added German translations
· using UniqueIds to add tracks to playlist (to preserve compatibility across versions of Amarok)