jamiebali / mrscc Goto Github PK

Each neurone in a HNN (Fully recurrent neural network) could represent a work (excluding stop works, punctuation, etc...) and co-occurance of words in sentences could lead to strengthened connection between words.

This works technically function similarly to an n-gram, but would be fuzzier. The question is, does a fuzzy HNN grant better accuracy - being that it has the capacity to look at the whole sentence at once, but ignores order.

Dynamically growing dictionaries aren't fun, so we should just make a tri and quadrigram. Would be easier, and work just as well.

We could then make a 5-gram and show that it starts to get worse with decreasing frequency of combinations.

An experiment here could show which n-gram size is the best. This isn't needed, but it's a good fun extension that shows knowledge, gives more writing points, and gives explanation.

It's just faster than python, and if I want to look through the whole dataset, it might be faster to make dictionaries and stuff in C (#/++) and then do the bulk processing and prediction in python after