Broadband World Forum is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 3099067.

Jay Shah , Chief Data & AI Officer at MyRepublic, on Artificial Intelligence in Telecom Companies

Anna Sliwon-Stewart, research analyst, security and building technologies at Omdia, caught up with Jay Shah, Chief Data & AI Officer at MyRepublic, during Broadband World Forum 2019 to discuss how telecom companies are using AI to support their networks.

Written by the Broadband World Forum Team

WATCH THE FULL INTERVIEW HERE:


Anna Sliwon-Steward: Hi. My name is Anna Sliwon-Steward, and I’m a research analyst with the Security and Building Technologies Team at Omdia. I’m here today with Jay Shah, Chief Data and Artificial Intelligence Officer at MyRepublic. Hi, Jay. How are you?

Jay Shah: Hi, Anna. I’m good. Good to be here. Thank you.

Anna Sliwon-Steward: Great to have you with us. I wanted to talk to you today about artificial intelligence. And this is by far the biggest buzzword in the industry, and no matter the industry that you look at. So, would you be able to say what are the biggest misconceptions that you’ve noticed about artificial intelligence in the telecommunications industry?

Jay Shah: That’s a very deep question. There’re so many ways to answer that. First is, is it really AI that telecom companies are actually using, or are they simply doing big data analytics and trying to mask that as AI? To truly build AI solutions takes time and requires an incredible amount of data sources. So, once you have everything together, then you start building models, but you have to do it in stages. A lot of the times, telcos are saying that they’re using AI, but it’s really just smart rules environments or analytics.

Anna Sliwon-Steward: So, we’re not quite there yet, you would say.

Jay Shah: Getting there. From a telco perspective, telcos have access to more data per consumer than even some of the biggest tech companies in the world. So, I think we’re the closest to possibly get something going, but it’s also about getting the right mindset of the right individuals on board in the company. So, depending on what you’re trying to build your artificial intelligence for. If it’s to do with profiling individuals or profiling customers, for example, then make sure your data science and data engineering team have some interest or background in philosophy or human psychology.

Anna Sliwon-Steward: So, when it comes to AI being applied in a network environment, what would you say are some of the quickest wins that telcos can cash in on by deploying solutions like that?

Jay Shah: There are a couple of ways. One, you could look at self-healing networks, from a cellular network, a mobility cell tower data. So, you’re optimising your routing tables within the SIM cards, so that the individual has the best experience of making phone calls, and he’s not connected to a cell router or a cell tower that’s packed with a lot of already simultaneous calls taking place at the same time. On a fibre or broadband service perspective, it’s analysing the data that’s coming through, or the ports that are being used on the households that you’re supplying to, and optimising traffic to the… For example, for us, we’re big on gaming. Gaming is a huge thing about our business, and we optimise that at the network level. So, as a consumer, you get the best experience, from the lowest latency, lowest amount of hops, and the best gaming experience, joining most of the gaming servers in the world. So, we’re using data modelling and smart routing, so to speak, to be able to optimise that experience for the consumers.

Anna Sliwon-Steward: So, it seems to me that in order to be able to deploy these solutions, you do need skills, you do need certain job roles that could help you be more effective with deploying these solutions. But what would you say are the specific job roles or skillsets that you would expect from anyone coming into this industry?

Jay Shah: Finding the right mindset. It’s a tough one, but the biggest problem with big data analytics is not the data itself. It’s the mindset of the individual looking at the data. So, having open-minded individuals that can actually see beyond what they’re currently been doing in their business environments. That’s very key. And if you’re building a data science and data engineering team, then depending on what functions you’re trying to build, you try to bring in the individuals that have an experience or background in philosophy and human psychology. At the end, if you’re trying to build profiles of individuals, you need to understand how people actually even think in the first place, before you start building models, and then expecting a different output and wondering what’s going wrong. So, it really comes to getting the right people on board. That’s very key. Out-of-the-box thinkers.

Anna Sliwon-Steward: Everyone listening to this video should really take pointers now, if they want to get involved in AI.

Jay Shah: Absolutely. Think out of the box.

Jay Shah on Using Data to Optimise Services

"On a fibre or broadband service perspective, it’s analysing the data that’s coming through, or the ports that are being used on the households that you’re supplying to, and optimising traffic to the… For example, for us, we’re big on gaming. Gaming is a huge thing about our business, and we optimise that at the network level. So, as a consumer, you get the best experience, from the lowest latency, lowest amount of hops, and the best gaming experience, joining most of the gaming servers in the world."

Anna Sliwon-Stewart: So, when it comes to MyRepublic’s activities, how’s your organisation harnessing the power of artificial intelligence?

Jay Shah: First it’s getting access to data. So, we built our own business support system and operational support system. We don’t have data siloes. We have very detailed data maps and inventory that we know exactly where all of our information sits with the server, who’s got access to it. We consolidate that into a massive data lake, and we create data cubes. Once we create data cubes, we can than start querying this information from multiple different departments. So, we’re looking at sales information, marketing, customer services, finance, billing, operations, and looking at all of these different… 

In a normal telco, it would be disparate data sources, but in ours, it’s all combined. And we’re identifying new patterns. From the patterns that we identify, we’re drawing insights. And when we draw insights, we then write reports for the various different business units, to suggest actions that they should perform.

Anna Sliwon-Stewart: Great. That sounds really interesting.

Jay Shah: It’s fun.

Anna Sliwon-Stewart: Okay, so we’re at a massive event today, Broadband World Forum, bringing everyone in the industry to talk latest innovations, latest solutions. So, what do you think about the forum so far?

Jay Shah: I like it. I love it. I’ve been here for three days, and great networking. It’s been awesome.

Anna Sliwon-Stewart: And I guess this is your first time, right?

Jay Shah: This is my first time to the Broadband World Forum, yes. And the networking’s been great. They keynotes have been fantastic. There is a lot of talk about 5G, of course, because 5G will definitely impact broadband at some point in the future, but I don’t think we’re there yet. So, if you’re broadband service providers, you don’t have to worry at present.

Anna Sliwon-Stewart: Sounds good. And briefly going back to AI in general as a massive trend, we’ve got 5G as a massive trend, IoT, and AI, so what do you think is the potential for AI to reshape the world we live in?

Jay Shah: You could do it in many different ways. The models that I’ve built is based on the eight factors that contribute to how people actually make decisions. So, looking at the psychological factors that contribute to how people make decisions, and using these models to be able to build profiles, and then we can build a different and unique type of experiences. But to give you an example, let’s look at smart homes. Everybody’s talking about smart homes. You need great fibre. You need a great Internet connection for these environments.

But if you’ve got a smart home, what does it mean? That you can just say, hey, Google switch the lights on? And that’s smart? Or, is it that it actually monitors your sleep patterns and temperature settings within every single room, and identifies what needs to be changed automatically, so that it optimises the entire experience, the sleep experience, for the individual. Now, you could do that using artificial intelligence. You can do that by combining these information sets and start building profiles of people.

Anna Sliwon-Stewart: That sounds very interesting, and I think we can all see how this could apply to everyone’s life.

Jay Shah: Absolutely.

Anna Sliwon-Stewart: And the minute you push it out to the industry, to higher levels, that can bring in massive changes.

Jay Shah: It will. Absolutely. And if you’re looking at an industry where people are thinking, how do I actually sell more broadband or 5G, the biggest problem in this area is in the entire telecommunication industry, is the retail prices are coming down, and they’re coming down really fast. Your costs still remain the same, or slightly going higher. Your margins are getting squeezed. So, you have to think out of the box. You have to think of ways to actually provide new experiences, new services to your consumer base. Otherwise, it’s a very slow death.

Anna Sliwon-Stewart: Industry, have you heard that? There’s a challenge for you. You need to address it. So, we looked at the overall perspective of how AI is really shaping our current world, but how is MyRepublic specifically contributing to that digital world?

Jay Shah: So, as we built our own business support system and operational support system, we’re not only using it within our network. We have other telcos actually now deploying our technology into their environments. So, we’re sharing our knowledge. We’re sharing our technology.

Anna Sliwon-Stewart: I guess this is the spirit with which we can maybe see artificial intelligence and other modern technologies spreading all around.

Jay Shah: Absolutely. Collaboration.

Anna Sliwon-Stewart: Great. Wonderful. Wonderful talking to you. Thank you so much for being here.

Jay Shah: Thank you, Anna.

Anna Sliwon-Stewart: This was Jay Shah, Chief Data and AI Officer at MyRepublic, and we’re talking at Broadband World Forum about the impact of artificial intelligence. Thank you.