Authors: Tofara Moyo
We introduce a type of cascade correlation network to predict the next word in a sentence, where the hidden layer is connected in such a way as to represent the topology of the conversation up to that point. There will be one input neuron that is fed the current word, and one output neuron that will output the predicted next word in the sentence. During training, we will be given a set of conversations to train on. That means that we shall know the input output pairs to train with, and the topology of the previous parts of the conversation correlating with each pair. We shall build the hidden layer in such a way that it is isomorphic to the previous parts of the conversation at the point here that input output pair coincide. During operation we will build the hidden layer of the network at each input word and in such a way as to be isomorphic to the previous parts of the conversation up to that point. Training will make the probability of the output n+1 word to be conditionally dependent on the previous words in that order and not just the nth word, conditional on the training conversations giving the chat-bot the ability to talk consistently on topics and themes found within the training conversations.
Comments: 2 Pages.
[v1] 2019-04-24 08:50:57
Unique-IP document downloads: 0 times
Vixra.org is a pre-print repository rather than a journal. Articles hosted may not yet have been verified by peer-review and should be treated as preliminary. In particular, anything that appears to include financial or legal advice or proposed medical treatments should be treated with due caution. Vixra.org will not be responsible for any consequences of actions that result from any form of use of any documents on this website.
Add your own feedback and questions here:
You are equally welcome to be positive or negative about any paper but please be polite. If you are being critical you must mention at least one specific error, otherwise your comment will be deleted as unhelpful.