Empowering Others to Use Data

Loading...

Del curso dictado por Johns Hopkins University

Building a Data Science Team

1977 calificaciones

Johns Hopkins University

Building a Data Science Team

1977 calificaciones

Curso 2 de 5 en SpecializationExecutive Data Science

Data science is a team sport. As a data science executive it is your job to recruit, organize, and manage the team to success. In this one-week course, we will cover how you can find the right people to fill out your data science team, how to organize them to give them the best chance to feel empowered and successful, and how to manage your team as it grows. This is a focused course designed to rapidly get you up to speed on the process of building and managing a data science team. Our goal was to make this as convenient as possible for you without sacrificing any essential content. We've left the technical information aside so that you can focus on managing your team and moving it forward. After completing this course you will know. 1. The different roles in the data science team including data scientist and data engineer 2. How the data science team relates to other teams in an organization 3. What are the expected qualifications of different data science team members 4. Relevant questions for interviewing data scientists 5. How to manage the onboarding process for the team 6. How to guide data science teams to success 7. How to encourage and empower data science teams Commitment: 1 week of study, 4-6 hours Course cover image by JaredZammit. Creative Commons BY-SA. https://flic.kr/p/5vuWZz

De la lección

Building a Data Science Team

Welcome to Building a Data Science Team! This course is one module, intended to be taken in one week. the course works best if you follow along with the material in the order it is presented. Each lecture consists of videos and reading materials and every lecture has a 5 question quiz. You need to get 4 out of 5 or better on the quiz to pass. Overall the quizzes are worth 17% of your grade each, with the exception of the last quiz, which is worth 15%. I'm excited to have you in the class and look forward to your contributions to the learning community. Click Discussions to see forums where you can discuss the course material with fellow students taking the class. Be sure to introduce yourself to everyone in the Meet and Greet forum.If you have questions about course content, please post them in the forums to get help from others in the course community. For technical problems with the Coursera platform, visit the Learner Help Center.Good luck as you get started, and I hope you enjoy the course! -Jeff

Embedded Teams vs. Dedicated Groups5:29

How Does Data Science Interact with Other Groups3:23

Empowering Others to Use Data5:02

Conoce a los instructores

Jeff Leek, PhD
Associate Professor, Biostatistics
Bloomberg School of Public Health
Brian Caffo, PhD
Professor, Biostatistics
Bloomberg School of Public Health
Roger D. Peng, PhD
Associate Professor, Biostatistics
Bloomberg School of Public Health

It's not just the people in your Data Science team that will want to

use data and interact with data.

So part of your role as Data Science manager, or

a Data Science executive is teaching people how to use data or

empowering people to use data within an organization.

So one of the ways in which to do that is through Data Science training.

So there's a number of different ways that you can do this.

One is pointing people to resources, collecting and

curating resources that are available online.

Whether that's online classes or online materials related to sorts of things that

they might wanna be able to do.

It might be writing those tutorials yourself, so

that people can interact with the data themselves directly.

Another really useful way is to have internal talks that are available to

everybody to go to if they want to.

Where the Data Science team communicates.

How do you pull data out of the database?

How do you make plots?

How do you visualize data in a way that's compelling?

How do you do machine learning?

Or what is machine learning?

Those internal talks are really useful if they can also communicate what

the internal capabilities are.

One thing that's very difficult about Data Science teams is that every Data Science

team had different capabilities.

Depending on the people that are there, how subscribed their activities are,

how difficult it is to do the problems.

And so communicating what's possible and what's not possible in a way that's sort

of unbiased to the whole organization often spurs new ideas.

New ways to take advantage of the Data Science team to solve new problems

that people care about.

So you can either have sort of training sessions or

sort of informational sessions about how the Data Science team works.

Those are both really useful ways to empower people to approach

the Data Science team and work with them.

Another way is to build documents or build tools or

build products that can be shared with other people.

Who don't necessarily have Data Science or Data Engineering training, but

would like to interact with the data.

So one common example is sort of a dashboard,

sort of a set of visualizations or set of plots.

That are built, so that somebody can click on some buttons and

interact with that visualization.

They can ask their own questions, have their own hypothesis and

interact with the data.

Even if it's just a subset of the data.

Another way is just one interactive visualization can be really compelling,

can be a way for people to sort of interact with the data,

ask one specific question.

And sort of see how their choices work.

So here the key is, there's a design element to it.

It's how are you going to design your interface so

that people can interact with the data.

They can ask their own questions.

But they're not intimidated by syntax or knowing how to actually do the stuff.

With the low-level programming languages that you or

your Data Science team might know about.

So in addition to training,

there's these sort of ways to directly connect people to the data themselves.

Ideally, you can automate this.

So a lot of report writing tools, whether they're in sort of high Python

notebooks or shiny documents or so forth, can be sort of auto-generated.

You can also setup a dashboard that does an auto-pull from your database,

that always has the latest and freshest data available to people.

If you can automate it in that way, the most common questions can be

answered by people other than the Data Science team.

Which leads them to innovate and to focus on new ideas and

new things that they can do well.

Or to shore up the infrastructure for this sort of Data Science processes that you

already have going on in your organization.

The other thing that can be really useful as a way to have people sort of interact

with the Data Science team is with Data Science idea evaluation.

So often people will have ideas about ways they think they might be able to

use data that solve a particular problem or another.

But it turns out that sometimes those are impossible.

We don't have the right data or we don't have the imaginary infrastructure,

we don't have the right scale to be able to solve that problem.

But it's still really nice if you can provide a way or a format or

a forum where people can propose ideas that they'd like to do with the data and

get critical, but positive feedback on those ideas.

You don't wanna immediately crush any idea that somebody brings up.

You'll never hear another idea again.

But it is a good idea to sort of, that helps tune expectations.

What can we actually do with the datasets that we have?

Especially, if someone else is proposing the ideas and

you are sort of letting them know.

Yes, we can do that.

No, we can't do that.

Yes, we can do that, but

it would take a whole new infrastructure on the server side.

Yes, we can do that, but

it would require us to develop a whole new machine learning infrastructure here.

That gives people an idea about what's going on with the Data Science team.

What can you actually do, what are the parameters that makes it useful for you.

And so you can do peer review of these ideas from the data team and

from yourself as another way to interact with the organization.

But in general, the idea is as often as possible drawing

other people into the Data Science process.

Often that requires a little bit of education and that education can feel like

wasted time if you're not thinking about it in the right way.

But if you think about it as an investment in the future Data Science

capabilities of your organization.

It can be a really positive, really value compounding way of doing data science.

Explora nuestro catálogo

Inscríbete de manera gratuita y obtén recomendaciones personalizadas, actualizaciones y ofertas.

Coursera brinda acceso universal a la mejor educación del mundo, al asociarse con las mejores universidades y organizaciones, para ofrecer cursos en línea.

© 2018 Coursera Inc. Todos los derechos reservados.

Descargar en la App Store

Consíguelo en Google Play