One of the few applications of Machine Learning (M.L.) –a computer that can improve it’s output through learning– that becomes Artificial Intelligence (A.I.) –a computer emulates human intelligence– is in ‘chatbots’, also called virtual agents, where humans can interact conversationally with a machine. Usually this is used in customer servicing although it can be used for retailing and non-commercial applications like therapy.
Creating a chatbot is surprisingly easy if you know now. What you need is:
- An A.I. Engine that can understand what your customers are asking
- To train your A.I. Engine using sample questions you have in your FAQs, social media interactions, and customer service logs
- A console like Zendesk, Twilio, Intercom or Salesforce so live agents can answer questions that A.I. cannot.
- An Instant Messenger like a pop-up windows on a website, Facebook, or WhatsApp
- Answers to the questions your customers ask