TURING - Text Universal Recognition In Natural Gateways

This was done as a school assignment (Applied Artificial Intelligence at Blekinge Institute of Technology).

The idea is to load two texts and save the characteristics. You then provide a sample from one of the texts. I then should be able to decide from which one of the two text your sample comes from.

Training set 1 | Training set 2 | Sample

Behind the scenes (Technical details)
Text Word count Name Word length Sentence length Commas Newlines Similiarity Semicolons words/ line Quote length Colons Words t1 Words t2
Text 1140180.84695.102832.52440.05730.331815490.10998.02812.97390
Text 2411410.96274.535215.35070.10627.539907.093178.92865
Text 325214.992131.50.06350.12515490.525226.80147
CLICK TO VIEW THE RESULT

Algorithm 1: One point for the closest match.
Algorithm 2: The assigned score is the ratio between the difference of the closest and not.

In favor of text 1: 11 -------- 135.1544

In favor of text 2: 0 -------- 0