Chatbot development resources from IBM, Microsoft and cloud providers for building a chatbot have exploded with comprehensive web services offered, and in 2017 there are many to explore.  This article is a summary of what a developer would explore.  This is not a definite list and there are no affiliations with any.  Reasons for chatbot development include:

  • Robot not programmed to do chatbot developmentIncreased operational efficiency (customer service)
  • Quicker self service product development (knowledge base / machine learning)
  • Notifications and automation
  • Interactive advertising opportunities
  • Games & … many more …

Conversational bots are the main area of chatbot development.  Natural language processing (NLP) and understanding are the main features for this type of product.  Further from NLP, there are speech and vision APIs to extend chatbot development for further immersion and engagement.

Before listing the serious chatbot stuff, there’s fun stuff like houndify which is handy for a fun banter chatbot between a group of friends.

Enterprise Chatbot Development

Microsoft has made cloud services a first class citizen and adding conversational intelligence to apps is right up there.

IBM offers Watson Development Cloud to enable cognitive computing featuring language, vision, speech and data APIs.

Cloud, Platform & Business Chatbot Development

The more accessible (i.e. less developer resources) avenue is to take cloud services for a spin.

Motion.ai makes chatbot development easy.  Great for prototyping.

Wit.ai makes NLP quite quick to develop models of entities like duration, location and an intent to gather information from a human and analyse the best route to take for servicing their needs.

Human Computer Interaction (HCI) and chatbot development

There is a great write up from a personal assistant chatbot organisation, how to teach a machine to understand us.

How much of the tech world will be affected by automation and chatbots.

A chatbot that saved time and money for the court system and residents of the city of Atlanta.

The types of chatbots on a platform like WeChat differ from the types of chatbots we find on facebook messenger or slack.

Nadia was a scary ‘human’ chatbot used by the Australian government to test human-like chatbots.

For engineering teams, some simple scripts can be used to tie disparate systems and people together.