How to use AI chatbots to connect with customers founder and CEO Eitan Sharon explains how AI chatbots use natural language processing to improve the customer experience.

How to use AI chatbots to connect with customers

AI chatbots using natural language improves customer experience, founder and CEO Eitan Sharon tells TechRepublic Senior Producer Dan Patterson. The following is an edited transcript of the interview.

Dan Patterson: Artificial intelligence is one of the hottest buzzwords in business technology, and the applications are nearly infinite. I wonder if you could help us understand, not just how AI is being used, but the technology behind AI chat bots?

Eitan Sharon: The ability to do that more thoroughly is supported a lot by the advances in the field of artificial intelligence and deep learning, which allows us to understand the visual and the language requests of the users much better.

Dan Patterson: So tell me a little about the technology at play. It's one thing for us to say AI, but as we all know there is predicted analytics, machine learning, natural language processing. How do AI based chat bots take advantage of AI technologies?

Eitan Sharon: So in our specific case,, we have put big emphasis on visual analysis of content. We believe that for things that have to do with clothing apparel, home design, interior design, the visual aspect is pretty much overwhelming. It's very difficult to describe clothing fashion and things you see in words, to find it, to match it up with other things, to show our opportunity and outfits and combinations to the users. That we basically use the artificial intelligence so far image analysis, so that we find on people's looks, street style, and others... what they wear, similar and relevant things, combinations.

SEE: IT leader's guide to the future of artificial intelligence (Tech Pro Research)

We are able to recommend to them in a personal way, how to improve and upgrade and refresh their inventory of clothing. That is in our case in combination with of course text. So we're talking chat box at the end of the day, and then people can take pictures, and interact with pictures, as well as use of course language to say, cheaper, more luxurious, different colors, and the combination of which starts a conversation that we follow, record this conversation on AI.

Dan Patterson: In what way do you interact with business partners - say, fashion brands or manufacturers -- to translate visual styles into language that AI can understand?

Eitan Sharon: So the common way we operate, is we work with partners such as Levi and Louis Vuitton, and others. We offer a channel for them to talk to the users through a bot, the kind we described. It would be a bot that is co-branded, maybe branded as the retailer, or the brand in our channel, offered on the site, and properties, as well as through Facebook messenger, and other venues. Their users can talk to the brand, and say, I'd like to buy something new or do you have I'm looking for things of that sort, and then there would be responses with images and offerings of inventory. They would click on stuff, and ask for more like that, and style-me please, and how to wear this, and, at the end of which, they can actually buy with basically power of business-to-business-to-consumer channel, for one-on-one interaction with the user, to, and with, the brand.

Also see