Intelligent Chatbot Authoring Platform using Sentiment Analysis & Neural Network

Intelligent Chatbot Authoring Platform using Sentiment
Analysis & Neural Network

The Challenge

Well before the advent of Generative AI, our client (in 2016), identified an opportunity to disrupt chatbot technology by building an intelligent chatbot authoring platform. Platforms for chatbot creation, at the time, required significant crafting from chatbot authors and had limited domain definition capabilities. The client wanted an engine that could create AI enabled chatbots, and train them using ML tools to enhance ability to develop a deeper understanding of the customer and thereby converse more intelligently.

Hypothesis

  • Retrieval based chatbots can be authored to incorporated various implicit signals (advanced language models) to be more effective when interacting with customers.
  • Using similar language traits to your customer can improve trust and comradery and help close sales quicker.

Execution

  • Our approach was to create a platform for chatbot creation that uses domain specific advance language models that allows the system to define entities and intents of the customer talking to the chatbot. This enables deeper understanding of the wants of the customer and helps the chatbot give an appropriate response.
  • Tools used included: Conditional Random Fields, Sentiment Analysis, Topic Modeling, Recurrent Neural Networks (RNN), LSTM.

Outcomes

  • A Chatbot authoring platform for curating chatbots to recognize what customer wants & respond in a tone and language most amenable to the customer.
  • Iterative approach to continuously improve the chatbot with a deeper ability to understand and converse with humans.

Project Highlights

38%

Increase in Chatbot engagement and Utilization