Question Answering Systems: How Search Engines Answer Questions Right Away

These days, search engines are often providing us with the exact answer we’re searching for thanks to technology known as the Question Answering System. This includes open systems, like Google, and closed systems, like a chatbot or an FAQ that utilizes deep learning technology to provide the most accurate information to users.

Tia explains how Question Answering Systems can help with everything from product support and financial services to medical and legal searches.

You can binge-watch all of season two or jump on back to watch season one.

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

Transcript:

(voice off camera) What was Prince’s real name!?

Prince Rogers Nelson!

Have you noticed that Google is getting straight to the point these days?

(looks back at phone and begins humming „Purple Rain“)

These days search engines are doing more than providing a list of articles for us to comb through. They are ready and rarin’ to give us the exact answer we’re searching for.

This is thanks to technology known as the question answering system, and there are several different types at work today.

On one hand there are open systems – like Google – that can theoretically answer any question you toss its way.

On the other hand is the closed system. These systems only know how to answer questions within a certain subject – but they answer them like an expert. Like when you talk to a travel agent chatbot. It can answer “How do I cancel my flight?” but not “How do I change the oil in my Toyota Corolla?”

FAQ systems are closed systems. They use deep learning in their searches to provide the most accurate information based on previously answered questions.

There’s also the knowledge-based question answering system which is built strictly on a set of facts the machine has learned.

And the knowledge graph question answering system keeps track of how all its data is related.

Now we don’t always ask perfect questions. In these instances, we need the computer to be smarter and make inferences. That’s where Deep Learning and Neural IR Search can come into play.

These Question Answering systems can help us in product support, financial services, and even legal or medical searches. Customers can find answers, employees are more productive, and trivia seekers can cut time seeking.

(voice off camera) What’s Madonna’s real name!?

Madonna Louise Ciccone!

(voice off camera) Wow! You sure have a lot of information!