Predicting with Signals in Artificial Intelligence Explained in 3 Minutes

Signals are the hints about their intentions that people leave behind. Lucid Thoughts explains what signals are and how they can be used in business to predict and influence customers.

You can binge-watch both  season one and season two right now.

Subscribe to the Lucid Thoughts channel and be sure to leave your questions and comments on each video.

Transcript:


Welcome back. Signals, you send them. You receive them.

We send a lot of signals in real life and online.

When tech folks talk about signals, they mean the things we do to help others learn about us. And everything we do gives off signals.

The things we click on. The things we don’t. The things we search, also known as our queries. The things we put in our carts and the things we leave in our carts. The things we eventually buy and even the things we hover over. They all give off signals.

One way we use these signals is to cluster and classify users. Remember that? Yeah. You can cluster and classify users based on their behavior to provide the right recommendations to the right users. We’ll have more on that later.

Another way to use these signals is for something called A/B testing. This is a method of putting two options out there to see how users are going to react to them.

Let’s take you back to the classroom.

Say Ms. Harris wants to see what videos her class respond to. She shows some of the kids a video that teaches math using kittens. And some kids see one that teaches math with puppies.

And after showing each video, she sees the kids that watch the puppy video were more excited about math and picked up the lessons quicker. Puppies win. Ms. Harris was A/B testing the kittens versus the puppies. And the signals her students gave off, told her, that the puppies were the way to go.

One more way we can use signals to give a boost to items on search engines.

Let’s say Ms. Harris is searching for more puppy videos to show her class. Because of the signals Ms. Harris has given off previously, searching for educational puppy videos, lesson plans relating to these puppy videos and puppy worksheets for kids, the search engine is able to provide more relevant results when Ms. Harris simply types in, “educational puppy videos”.

Every signal sent is a piece of JavaScript Object Notation, little bits of data. With each person giving off thousands of these tiny bits of data, if you try to read all of them, your head just might explode.

That’s why it makes sense to bring all of these signals together to create your digital footprint. Then we can follow these footprints to make sure the users are getting the experiences they’re looking for. Your customers are sending you all the right signals, and now you’re ready to listen.

Catch you later.


Related Videos
Virtual Assistants: How Question Answering Systems Help Siri, Alexa & Google Home
Question Answering Systems: How Search Engines Answer Questions Right Away
Query Intent Classification: How Does the Internet Know What I Mean?