I intend to create various neural networks that will help categorize poems after their authors. In the first phase I’m going to use letter patterns to do that, as per a paper by Johan F. Hoorn, Stefan L. Frank, Wojtek Kowalczyk and Floor Van Der Ham. Unlike in their research, I want to use self organizing maps, not supervised feed-forward networks.
In a second phase, I’ll replace the letter combination input with some NLP analysis, though exactly how I still have to document myself.
Practical purposes of this application would be to categorize popular poems (creations with multiple authors - folklore), to help in settling authorship disputes etc. In future, I want to expand this to prose and even music.
The development will be done in Java 5.
These details are provided for information only. No information here is legal advice and should not be used as such.