SEMANTRIS

What is Semantris ?

Semantris (developed by Google AI) is a set of word association games powered by machine learned, natural language understanding technology. Each time you enter a clue, the AI looks at all the words in play and chooses the ones it thinks are most related. Because the AI was trained on billions of examples of conversational text that span a large variety of topics, it’s able to make many types of associations.

Developed by:

Ben Pietrzak, RJ Mical, Steve Pucci, Maria Voitovich, Mo Adeleye, Diana Huang, Catherine McCurry, Tomomi Sohn, and Connor Moore.

How it works:

How does a computer understand when you talk to it using everyday language?

Google's approach was ton use billions of lines of dialogue to teach an AI how real human conversations flow.

Once the AI has learned from the data, it is then able to predict how likely one statement would follow another as a response. In these demos AI is simply considering what you typed to be an opening statement and looking across a pool of many possible responses to find the ones that would most likely follow.
The technique that Google uses to teach computers language is called machine learning.

Input data: The input data is a billion pairs of statement, where the second statement is a response to the first one.

Predicting: Google is predicting the response to a question or a statement. After seeing all those pair of sentences and responses, the AI learns to identify what a good response might look like.

Model: The trained system that is used for making predictions. After training, Google's  model is able to pick the most likely response from a pool of options.

Check out, Semantris.

Source: Google