Machine Learning (ML) is a way for a chatbot to learn from data and make decisions without being explicitly told what to do.  

It uses algorithms to understand the words in a conversation and figure out how to respond. Every time it interacts with someone, it stores what it has learnt so it can use it in future conversations.  

The more it converses, the more it learns. If it ‘fails’, it uses that learning to improve its understanding of language. 

The basic Machine Learning categories are supervised and unsupervised learning: 

Supervised learning is a type of Machine Learning algorithm which uses labeled data to train models. This data is labeled in a set of categories, which the algorithm then uses to make predictions about new data/inputs.  

Unsupervised learning is a Machine Learning algorithm which does not need labeled data. These algorithms are used to find patterns and relationships in data and to discover insights to make decisions. 

Machine Learning is becoming increasingly powerful and the level of accuracy is growing all the time. 

For more detail, read "Machine Learning Simplified"

Learn more about: 

• AI chatbots 
• Conversational AI 
• Neural Linguistics 
• Named Entity Recognition 
• Sentiment Analysis