Identifying authors through a computer program.

[whitesa]

I am an eighth grader and working on background research for the project idea of building a computer app that identify's author by their writing. I was wondering if you had any recommendations for sources.

[hhemken]

whitesa,

Have you looked at the ScienceBuddies version?:
http://www.sciencebuddies.org/science-fair-projects/project_ideas/CompSci_p022.shtml

The trick is to use counts of various things as described there to distinguish between authors. For more ideas, make sure you google something like this:

Code: Select all

text analysis identifying authors

For a huge sample of texts, try Project Gutenberg:
http://www.gutenberg.org/
I would recommend you use the plain text versions. You would have to cut out the extraneous stuff at the beginning and the end of texts.

If you can run your program against many large texts, you may also be able to classify them by rough date of publication, author gender, and who knows what else.

Good luck!