A Deep learning based chatbot based on Neural Conversation Model. This Project is a chatbot powered by deep learning and trained on data from Cornell built on TensorFlow v1.5.1and Python v3.6.4 and is based on conversational modelling. Conversational modelling is one of the most exciting problems in the field of Natural Language Processing. Our approach is based on producing answers given by a probabilistic approach to maximize the correctness of answer in the given context.
#SOFTWARE AND TOOLS NLTK Tensorflow Seq2Seq model
#Datasets We trained our model on the OpenSubtitles dataset by Tiedemann . This dataset consists of movie conversations in XML format.
#Conversation:
Q: Hello
A: How are you today?
Q: What is your name?
A: Bedevere, my liege.
Q: What is your favourite number?
A: 105-353.
Q: What do you like?
A: I do n't know.
Q: You look sad.
A: I was just having a good one.
Q: You are funny.
A: Me, too, i'm sorry.
Q: No, No. It's fine
A: There are no charges, though.
Q: What is your favourite sports?
A: To seek the grail.
Q: Are you a poet?
A: No, i'm divorced.
Q: What is your father's name?
A: `` dad.''
Q: Are you an orphan?
A: Uh huh.
Q: When did your father die?
A: Last night.