Source – bgr.in
Today, Artificial Intelligence (AI) and Machine Learning (ML) are two popular terms that tech companies cannot stop talking about. Everyone from Google and Microsoft to Apple, Samsung and Amazon are going big on AI. Besides smartphones, smart speakers, voice assistants, apps, connected cars, security surveillance, healthcare and customer support are other areas where AI is used.
Machine Learning (and deep learning) has been going on for years, and now with the data that exists, tech companies are putting it to the best use. On device machine learning combined with artificial intelligence can help in anticipating things in advance. But while you may not know, AI and ML can be very helpful in customer service too. In fact, according to a recent research by Gartner, 85 percent of client relationship with a business will be managed without human interaction by 2020.
Now, in a conversation with BGR India, Olivier Klein, Head of Emerging Technologies, Asia-Pacific at Amazon Web Services, offered some insights on how the company and its clients are using AI and ML. He also mentioned that Amazon has been doing machine learning for the past 20 years, and now, the company has made it available for end users through the platform. We spoke about how AI is evolving to be the core, and at the forefront of improving customer experience, and Olivier offers a number of examples to help understand the benefits.
24-hour customer support, Chatbots
There are times when you order some product from an e-commerce website, and you want to know the status of your package, or raise a complaint for refund or replacement in case of a faulty product. These are the times when we end up calling the customer support holding the phone for at least 5 minutes to get to the representative, just so you can explain your issue and ask for refund. But this process can be completely automated, and a lot of companies are using this to their advantage. And with several advances in natural language processing (NLP), these chatbots are getting better all the time.
Take an example of Amazon, when you click on ‘contact us’ you get the chatbot interface where you select your problem from the list and go forward. Most times, you don’t have to talk to a customer care as things are fed in the system and you just need to choose the relevant option. I recently had to cancel a product that didn’t get delivered in the said time, and the process was completely fluid. I got the product cancelled and refund processed in under two minutes without having to talk to a representative.
Facial analysis – Travel without passport and ticket
AI is today used for facial analysis to recognize and authenticate you. It is most commonly used in a lot of smartphones today, in the form of Face Unlock feature. But, there are a wide range of applications where facial analysis can be used. A good example could be while boarding an aircraft. Generally, you have to show your passport and ticket, but facial analysis can offer a seamless experience.
Your data, such as facial structure and other characteristics are already present in the system. When you head towards the check-in counter, the computer just has to run your face through the database, find a match, and then look for bookings that you may have made. Once authenticated, it can check you in, and offer seamless experience. Olivier mentioned that the company has been testing this solution in Asia Pacific region, but it may take a few years before it is deployed for use by mass audience.
Movies and series recommendation using computer vision
If you have an account with Netflix or Amazon Prime Video, you may have noticed that these streaming services recommend you top shows to watch. Ever wondered how the list is populated? Well, it uses machine learning model to understand your likes and recommend a TV show or movie accordingly. The Amazon AWS machine learning model runs facial recognition on top of the video to understand the actors in the video content that you are watching.
Along with facial recognition, computer vision model is also used to understand the scenes – is it an action packed movie with lots of action, chasing cars and explosions, and more. Based on this analysis, the computer vision and machine learning model understands that kind of things you generally like, and will offer content suggestions and recommendations based on that.
Amazon X-Ray and image recognition
When watching a TV show or a movie, have you even come across a situation where you feel that you have seen that face or actor before? At times it drives you nuts when you can’t remember where you’ve seen the face. This is where X-Ray feature comes to rescue. If you have Amazon Prime Video subscription, chances are that you may be aware of the X-Ray feature.
X-Ray is an image recognition service where the platform uses Amazon Rekognition API for image an video analysis. It also uses a feature called Celebrity Rekognition that runs on top of video to detect and recognize faces in the video. Once the recognition is complete, it then ties it back to IMDB. As you can see in the above screenshot, you can exactly see the actors in the scene along with their name and face. Clicking on the link will take you to IMDB page of the actor, where you can check all of their details.
Automatic due date reminders
Have you ever got an automated call from your insurance provider or your telecom operator to remind you about the payment due date? Well, these are automated calls and some of them use Amazon Polly services which allows companies to render a personalized and natural sounding voice response.
Policybazaar is one of the insurance service providers using the service. When the required details are feeded to the system, it will call the customer, greet with their name, read out the policy number and inform about the due date and the premium they need to pay. And the beauty is the voice is more natural sounding, than being robotic, offering a personalized experience.
Flight price tracking
Services like Google flights tracker, Skyscanner, Ixigo have deployed Artificial Intelligence that can help in a number of ways. Say you are looking for flight tickets from Mumbai to Bali, and the pricing shows as Rs 35,000, and you know anything above Rs 28,000 is a little on the expensive side. You can then set price alerts which will track the price changes on a daily basis and mail you about same. You can then see how the price is fluctuating, and book when you think it can’t go any lower than that.
Flight delays, train ticket confirmation prediction
AI also keeps a track of the flight timings, the delays in take-offs and early landings, and based on some permutations and combinations, it can predict if your flight will be delayed or will be on time. Same goes with train ticket bookings too. IRCTC’s new website, and its booking confirmation prediction system is one of the best examples.
The AI has years of data to track the waiting list tickets from source station to the destination for a given time. For instance, during peak seasons, Mumbai – Madgaon route is quite busy, and even a ticket with waiting list 15 seems difficult to get confirmed. Other times, even waiting list 80 gets confirmed. Based on the historical data, the system can predict the odds of your waitlisted ticket getting confirmed, thus helping you keep your option for other transport modes open at the same.
Home security and sentiment analysis
Artificial Intelligence can also help in home security where you simply need a CCTV camera and Amazon DeepLens API. Of course, you have to first feed to the system with photos of your family members, closed ones and the regular people, like the watchman, milkman, and more.
You will have to teach the Amazon Facial Rekognition API this is the face of my mother, sister, father and so on. It can also do facial analysis on the basis of whether you have a beard, moustache, etc. What’s more, the API can also read emotions and do sentiment analysis. It will use facial recognition tech to understand if it is the person you know, and if you don’t know the person, it can immediately text or call you.
Most of these solutions are already available out there and with Amazon AWS, you can either built them yourself or opt for professional help in for coding, testing and deployment. Services like Amazon SageMaker, Amazon Kinesis and Amazon S3 allow you to build a platform customized to your needs.