MetadataShow full item record
Chatbots or smart conversational chatting machines are being built using Artificial Intelligence and machine learning technologies to solve the existing problems in the area of natural language processing. Thanks to the recent advancements in the 'Human-computer interaction' field where text-based chatbot frameworks are the hot topic of research for the industrial sector as well as academia, initiated the development of advanced intelligent conversational chatbot systems. A new breed of these advance intelligent chatbots can provide much more versatility and agility to handle complex business or personal tasks quickly and more efficiently. In the last decade, Impeccable innovations in the deep learning field have forced the migration of template-based chatbot frameworks to the innovative trainable, lenient end to end based ML frameworks, thanks to the introduction of the recurrent encoder-decoder model  and recurrent neural networks (RNN). To overcome the existing limitation in rule-based or retrieval-based models during dialogue generation industry's favorite is the modern-day Sequence to Sequence-based framework with encoder-decoder-based architecture relying on attention mechanism for dialogue modeling including question answer-based conversations. In this thesis, we will provide an analytical overview of chatbot technologies, how it started and how much they progressed and what approaches were adopted throughout this time span. Later we will conduct some of the comparisons with sequence-to-sequence based on existing chatbot models based on different neural networks and compare the results for the audience.