.Net Foundation On .NET, 31 Mar 2020

Previous Meeting Next Meeting

⏯

youtube image

►

From YouTube: Making the case for AI, chat bots, and Power BI

Description

Microsoft has teams distributed across the world that are helping our customers solve problems that are very unique to their business and their communities.

While we were in the Dominican Republic, we sat down with Cloud Solution Architect Mabel Gerónimo from the local Microsoft office to talk about how they're combining tools like Bot Framework, Power BI, and cognitive services to solve interesting business problems in the country.

[01:13] - The value of Bot framework
[02:03] - Power BI Desktop
[03:10] - Using QnA maker to interact with bots
[04:52] - Triggering the bot

Bot Framework
https://dev.botframework.com/?WT.mc_id=ondotnet-youtube-cephilli

QnA Maker
https://www.qnamaker.ai/?WT.mc_id=ondotnet-youtube-cephilli

What is PowerBI
https://docs.microsoft.com/en-us/power-bi/fundamentals/power-bi-overview?WT.mc_id=ondotnet-youtube-cephilli

#PowerBI #Bots #MachineLearning #Azure #Dotnet

A

A

Hey everyone we're here at the Caribbean Developers Conference here in Punta Cana in the Dominican Republic, and you know, I, wanted to check in with some of our friends here the local Microsoft Office, to learn about some of the projects that they're working on. So why don't you tell me like who you are and what is that Claes is that you do here. Thank.

B

You Cecile, thank you for having me my name is marvel Geronimo I'm, the CEO say cloud solution, marketer for the local office here in Microsoft, America, Republic, yeah,.

A

B

Something that I've been working on, because most of my customer recently have been reaching out to get understanding and get insider projects about bot framework. Yeah.

A

B

Having the connection of that framer with power bi, which is also a familiar tool for most of those customers, gotcha.

A

Gotcha, so tell me: what is it like? Why is it that you said that you wanted to use bot forever and what what are you doing with it exactly yeah.

B

So more of our customers are currently realizing the value having an automated service agent in bot. Framer came as the perfect solution for Dan. You know to otherwise the process of resolving the doubts and questions that their customer may have been.

A

B

When we show them the bot framework, that is using.net, which is also a common language for most of the customers, yeah.

A

B

Was like the right tool for them to get started ahead very quickly, very easy, using tools such as Q&A maker. That also makes still.

A

B

Low and also you know like having the the integration with all of the rest of the tools in the Microsoft world in such as Microsoft teams and Microsoft power, bi, okay,.

A

Yeah that sounds really cool. Why don't you show us really quickly like some of the things that you've been working on yeah.

B

Yeah, so the idea for this, our report that you see right here, is that it's connected to the bot framer analytics that is currently being sent to the application insight service, not sure, okay,.

A

B

All of that and frustration is done with logic, ops, so.

A

B

Visual interface, when you can like define that the currency that you this to happen, how often you do you want this to happen? Where does this data come from and what type of data do you want to receive? Send that maybe put it into some continuity services? You know to analyze and key.

A

B

And then send that to a blob storage and then we'll fit into the power bi report that you see right here. So we get the location you.

A

Get the sentiments.

B

We get the confidence that the body's showing when it's responding, the answers with.

A

B

Words that the customers are looking for, so we get all the data that business people we need to to provide a decision. Sure.

A

What I love about this solution is just using so many different products and like putting them together like putting the puzzle pieces in the right place like make the right solution.

B

Yeah, that's correctly and the way to put those pieces together is even easier because he's old, visual interface right.

A

Right so we have logic apps in here we have application insights in here.

B

A

Services uh so you're doing sentiment analysis. So there's a little bit of AI machine learning- that's goodness insider here! So why don't you tell us a little bit about the box, so how do I tie family business like how do we interact with the button like what kind of information do I give it like? What does it do for me, yeah.

B

Yeah for this box right here that we had we, we use the Q&A maker framework, okay,.

A

B

Just a matter of going into this building already site that the Microsoft has entering the question and entering the answers that you may want it to handle first and then you can like plug in that shot, but that you create into your web page or.

A

B

A social media platform such as Facebook.

A

B

Know so your customers can have different channels to interact with the services that you're, providing as a as a company right.

A

And so with the Q&A maker I come to a website and I train it with these different questions and answer like these responses, yeah.

B

A

Over time, it becomes better and better because the more questions that we give it, the.

B

A

It becomes at answering those questions perfectly.

B

And the thing the cool thing about this is that you also get the source code behind the project say.

A

B

You want to go deeper, more advanced and incurable.

A

B

Necessarily are like what Q&A maker provides you, then you can do so as well. That's.

A

Awesome so that means that okay, so I have a bot. My business is using a spot right and then this information goes into Q&A maker answers questions whatever the case is so now. What's the thing that triggers the logic app to start running, yeah.

B

So in this logic, app I have a recurrence time of like trigger okay.

A

B

I set it up to every five minute in my case, I can.

A

B

That every every hour or daily, you know, depending on how much sensitive, how much time simcity is the data that I'm collecting right.

A

Right right, okay, so tell me like cuz again we can talk about technology. All we want right, but I think what's important. Here is talking about like how this technology is really helping. Empower the local people here in the Dominican Republic to just do work more efficiently. So how is this particular solution of this combination of tools being able to do that? I think.

B

That the way that these tools are combined and easier to being applied, you know that's just a way for people, so it doesn't require very exclusive technical skills they're only the technical people may have, but everybody can do it basically. So.

A

B

A window and also having this information, because sometimes you automate the process of delivering the service to your customers, but you don't want to lose in sight of that. You.

A

B

This information that allow you to get the insight not only from the customer perspective, but also from the chapbook perspective, so you can also know what to do better, what to improve in terms of showing things new things to learn to the checkbook. So having those two things together, its major win for the customer, and it is already like so easy to plug in then it's even better. So.

A

For people that are here in the Dominican, do you see a lot of ass for chatbots and these types of tools so like helping business, automation, yeah.

B

Yeah, like the the amount of customers that are getting interested in in this kind of projects, have raised increasingly its potentially yeah, so there's even more customers that are coming in the line. You know, and you know, starting to feel their customers, maybe in the same industry they have taken the step ahead and they're started to stay. I want that I need that, because it's an optimization of my operations right.

A

And so after we've gotten this information from the chat bot, it goes through our logic, apps pipeline and then I want to talk a little bit about the sentiment. Analysis that you use it.

B

A

It's I'm guessing it's checking like is this a positive statement? Is this a negative statement? So now you know using that API using cognitive services, we have that information that we can start making decisions about how efficient the chat box is.

B

A

Like is it answering the right question, or is it giving me an answer that makes me feel satisfied right? Those are the types of things that you're doing with it. Yeah.

B

Yeah correct, and not only to know that if my customer is happy about my child, but also if my customer is happy about my own services, you.

A

Know yeah because.

B

The more I get this kind of information that, in the past wasn't just as easy to get from. You know to yet available the more that I can understand my customer and their needs.

A

And it's an insight is so much important right.

B

Because now you.

A

Know that are we delivering the right tools and services, and are we actually helping or just being a complication to you know whatever? Is that the Briggs's process yeah.

B

Yeah yeah and making those questions is part of the value, as well of all, of the solutions. Cool.

A

This is awesome, man. Is there anything else that you'd love to tell us about this anything I.

B

Think just dad that you got it all I think that this is just a sample view of what you can get you.

A

B

Get more complicated than this, you can get more advanced. You know. At the end of the day, you are asked, you know the lead of the project. Yeah.

A

B

Are the one that decides what comes into the report but having already everything being connected? That's a plus that you don't even have to worry about awesome.

A

I, like I, got to come back it down here again another time and spend some more time with you guys in the office and see what y'all are working on well.

B

Calm sure thank.

A

You so much thank you for being on and thank you all for checking out this other episode, while we're here in the Dominican Republic and the Caribbean Developers Conference.

A