Hi all. I'm a newbie in ML. I read and found a paper about A Multi-Level Plagiarism Detection System Based on Deep Learning Algorithms and want to implement this model . But I can't find more about step-by-step guide to build it. How LSTM can make representation with input is list vector of sentence trained by Doc2vec.
Inputs of LSTMs are vectors and it does really matter what the vectors are, in signal processing there are signal windows, in NLP, these are usually word embeddings, here, they are document representations obtained from Doc2vec. Judging from the scheme, they probably use Doc2vec to obtain sentence embeddings. It is rather unusual, but definitely possible.
If the input vectors carry enough information, LSTM will certainly learn a reasonable representation.
