![\"headshot](\"https:\/\/www.microsoft.com\/en-us\/research\/wp-content\/uploads\/2019\/10\/Riham-Mansour-_Podcast__Site_10_2019_1400x788-1024x576.png\")
<\/h3>\n<\/h3>\nMachine learning is a powerful tool that enables conversational agents to provide general question-answer services. But in domains with more specific taxonomies \u2013 or simply for requests that are longer and more complicated than \u201cPlay Baby Shark\u201d \u2013 custom conversational AI has long been the province of large enterprises with big budgets. But not for long, thanks to the work of Dr. Riham Mansour<\/a>, a Principal Software Engineering Manager for Microsoft\u2019s Language Understanding Service, or LUIS. She and her colleagues are using the emerging science of machine teaching to help domain experts build bespoke AI models with little data and no machine learning expertise.<\/span><\/p>\n On today\u2019s podcast, Dr. Mansour gives us a brief history of conversational machines at Microsoft; tells us all about LUIS, one of the first Microsoft products to deploy machine teaching concepts in real world verticals; and explains how an unlikely combination of engineering skills, science skills, entrepreneurial skills \u2013 and not taking no for an answer \u2013 helped make automated customer engagement and business functions more powerful, more accessible and more intelligent!<\/span><\/p>\n Riham Mansour:<\/span> <\/span>I don\u2019t really care where I start in the research spectrum. I might start completely solving the problem from scratch, I might start by capitalizing on other people\u2019s work, I might put pieces together to make it work\u2026 The important goal to me is to have something that works for the purpose of the customer.<\/span> <\/span><\/p>\n Host: <\/span><\/b>You\u2019re listening to the Microsoft Research Podcast, a show that brings you closer to the cutting-edge of technology research and the scientists behind it. I\u2019m your host, Gretchen Huizinga.<\/span><\/b> <\/span><\/p>\n Host: Machine learning is a powerful tool that enables conversational agents to provide general question-answer services. But in domains with more specific taxonomies \u2013 or simply for requests that are longer and more complicated than \u201cPlay Baby Shark\u201d \u2013 custom conversational AI has long been the province of large enterprises with big budgets. But not for long, thanks to the work of Dr. Riham Mansour, a Principal Software Engineering Manager for Microsoft\u2019s Language Understanding Service, or LUIS. She and her colleagues are using the emerging science of machine teaching to help domain experts build bespoke AI models with little data and no machine learning expertise.<\/span><\/b> <\/span><\/p>\n On today\u2019s podcast, Dr. Mansour gives us a brief history of conversational machines at Microsoft; tells us all about LUIS, one of the first Microsoft products to deploy machine teaching concepts in real world verticals; and explains how an unlikely combination of engineering skills, science skills, entrepreneurial skills \u2013 and not taking no for an answer \u2013 helped make automated customer engagement and business functions more powerful, more accessible and more intelligent!<\/span><\/b> <\/span><\/b>That and much more on this episode of the Microsoft Research Podcast.<\/span><\/b> <\/span><\/p>\n Host: Riham Mansour, welcome to the podcast.<\/span><\/b> <\/span><\/p>\n Riham Mansour: Thank you.<\/span> <\/span><\/p>\n Host:<\/span><\/b> <\/span><\/b>I love situating my guests at the beginning of every podcast<\/span><\/b>. It\u2019s <\/span><\/b>such a research-y term. So let\u2019s situate you: you\u2019re a Principal Software Engineering Manager and you run the Language Understanding Service, aka LUIS. And I\u2019m absolutely going to have you tell us about LUIS in <\/span><\/b>a bit<\/span><\/b>, <\/span><\/b>but let\u2019s start with the macro view of who you are and what you do. What big problems are you trying to address and what gets you up in the morning?<\/span><\/b> <\/span><\/p>\n Riham Mansour: <\/span>Hmmm. <\/span>I\u2019v<\/span>e<\/span> started that effort of the Language Understanding Service with the Machine Teaching Innovation Group almost five years ago. And we\u2019re trying to solve a very interesting problem which is, how can we get people<\/span>,<\/span> and domain experts<\/span>,<\/span> who don\u2019t have a lot of machine learning expertise or even don\u2019t have data<\/span>,<\/span> or don\u2019t know what are the pieces that they need to put together to build an AI model<\/span>.<\/span> <\/span>S<\/span>o we\u2019re trying to get those people <\/span>on board to the AI world<\/span>\u2026<\/span> <\/span><\/p>\n Host: Okay.<\/span><\/b> <\/span><\/p>\n Riham Mansour:<\/span> <\/span>\u2026<\/span>because those people have lots of problems that AI can help <\/span>them with<\/span>. <\/span>So<\/span>,<\/span> we\u2019re trying to meet<\/span> <\/span>those people where they are today, helping them with AI<\/span> <\/span>solutions<\/span>,<\/span> powerful solutions to their problems<\/span>.<\/span> <\/span>In the past, AI was only coming from the giant tech companies that can afford having the AI experts and they have enough data, but <\/span>then<\/span>,<\/span> like<\/span>,<\/span> <\/span>now, enterprises and developers who want to build intelligent systems, how can they unlock the power of AI? I think this is the exact problem we\u2019re trying to solve.<\/span> <\/span><\/p>\n Host: Before we get technical<\/span><\/b>,<\/span><\/b> I want to talk a little more about you<\/span><\/b>,<\/span><\/b> because you bring what I would call <\/span><\/b>\u2018<\/span><\/b>hybrid vigor<\/span><\/b>\u2019<\/span><\/b> to the research party. You have roots across the tech spectrum from academic research to applied research to product and engineering\u2026 <\/span><\/b>And not necessarily in that order<\/span><\/b>!<\/span><\/b> <\/span><\/b>So tell us a bit about how all these parts make up your unified whole<\/span><\/b>,<\/span><\/b> and how it speaks to the work you do.<\/span><\/b> <\/span><\/p>\n Riham Mansour: <\/span>Yeah, s<\/span>o, actually, <\/span>I didn\u2019t mean to be all over the place! I think the journey of my career has <\/span>pushed <\/span>me into starting in engineering <\/span>and, back in the time<\/span>,<\/span> I was<\/span> <\/span>very interested <\/span>in doing software development, coding, and design<\/span>,<\/span> and all that. That was my passion back then and then working in the product brings a lot of rigor and you see all the real world problems and what can be done, what cannot be done and so forth. But all throughout my career in engineering<\/span>,<\/span> I always<\/span> <\/span>had the science appetite. I like to clarify things and play with the ambiguous. And being an engineer has part of that because you need to figure things out<\/span>,<\/span> but there\u2019s more to the dimension of science than just being an engineer because you start with a bunch of hypotheses<\/span>,<\/span> <\/span>w<\/span>here<\/span>,<\/span> like<\/span>,<\/span> we know nothing about how to solve that problem, and then we start tackling the problem through the scientific method. So I always knew, during my engineering career, that at one point I want to explore the science aspect of life and, when the right time came, I joined <\/span>a<\/span> PhD program. But then<\/span>,<\/span> when I was done with my PhD, the question<\/span> came<\/span> to my mind, what should I be doing? <\/span>And then, <\/span>I worked a little bit as a professor because this is the thing I didn\u2019t do before my PhD. I did the engineering part. So I tried to explore options<\/span>.<\/span> <\/span>T<\/span>hat was not a job for me<\/span>!<\/span> <\/span>Uh<\/span>\u2026<\/span> I<\/span>, l<\/span>ike<\/span>, t<\/span>he job of a professor is to make a delta change in the student life<\/span>,<\/span> and I love working at a fast pace with the smartest people in the world. So, this is why<\/span>,<\/span> like<\/span>,<\/span> a professor job was not for me<\/span>!<\/span> <\/span><\/p>\n Host: Move me to a faster lane.<\/span><\/b> <\/span><\/p>\n Riham Mansour: Exactly, get me to a faster lane<\/span>!<\/span> So I joined Microsoft Research and I started exploring that sweet spot for me and that\u2019s exactly where I found my real passion. So yes, engineering\u2019s great, academic research is awesome, but what\u2019s more awesome for me was, how can I build something that solves a scientific problem but yet<\/span>,<\/span> it\u2019s useful to people?<\/span> <\/span><\/p>\n Host: You joined Microsoft Research in 2012 and you came as a senior NLP applied researcher\u2026<\/span><\/b> <\/span><\/p>\n Riham Mansour: Yes.<\/span> <\/span><\/p>\n Host: \u2026in the Advanced Technology Lab in Cairo.<\/span><\/b> <\/span><\/p>\n Riham Mansour: Yes.<\/span> <\/span><\/p>\n Host: And you worked on several language innovations that ended up actually shipping in products. So tell us a bit, in turn, about the work you did in sentiment analysis, topic detection and key phrase extraction, because those are some of your big innovations.<\/span><\/b> <\/span><\/p>\n Riham Mansour: Yes<\/span>. Back in the time,<\/span> when I joined the Advanced Technology Lab in Cairo, my mission was to focus on making an impact out of whatever innovations that we work on. So<\/span>,<\/span> I always start with the customer. I don\u2019t like to start with the problem that interests me, it has to be a problem that interests me and solves a problem. So at first I went and<\/span>,<\/span> like<\/span>,<\/span> talked to multiple groups who are either doing NLP at the moment, or have plans to do NLP like the <\/span>Azure<\/span>Related:<\/h3>\n
\n
\nTranscript<\/h3>\n