0253 - Teaching Analytics Visually with Stefan Helfrich

https://is1-ssl.mzstatic.com/image/thumb/Podcasts211/v4/34/e4/ba/34e4ba17-7805-9145-6303-a991663adebd/mza_18028444146840710762.jpeg/600x600bb.jpg

Design Talk (dot IE)

Allen Higgins

184 episodes

1 day ago

Design Talk. A podcast for learning about the business of technology design and management. Listen to stories, panels, interviews and discussions about technology and design in-the-wild: the good, the beautiful, and the useful. Whether you've got one big idea, you think of yourself as an organisation designer, product designer, creator, or entrepreneur. Each episode offers a take on how people design, strategise, organise, and develop technology. We want to dig into the essence of design, discover the backstory to technologies, and unpack the design attitude. We started this podcast for you, because you are interested in tinkering, in making, and how ideas become 'things', and because we are all, in some way, involved in designing.

Hosted on Acast. See acast.com/privacy for more information.

All content for Design Talk (dot IE) is the property of Allen Higgins and is served directly from their servers with no modification, redirects, or rehosting. The podcast is not affiliated with or endorsed by Podjoint in any way.

Hosted on Acast. See acast.com/privacy for more information.

0253 - Teaching Analytics Visually with Stefan Helfrich

Design Talk (dot IE)

47 minutes 1 second

5 months ago

0253 - Teaching Analytics Visually with Stefan Helfrich

Hosted by Christina Philips and Anabela Da Silva Filipe Soares.

Welcome to today’s seminar by Stefan Helfrich.

In this session Stefan talks about the education paths on offer for data analytics and the need for balance between learning concepts versus hands-on experiences with tools. Stefan makes the case for the value of visual workflow approaches for teaching and implementing analytics.

How do we do that? KNIME implements a well-documented, comprehensive and capable software environment that enables users to design and operate data analytics workflows visually using the following objects:

· Nodes perform tasks on data. Nodes have inputs and outputs. Nodes have status/indicators. Nodes are natively implemented in Java. Python scripts may also be used as code nodes.

· Connectors link nodes. Connectors indicate data flows. Connectors send data from one node to another. Connectors have direction. Nodes plus connectors enable you to create workflows.

· Workflows are designed aggregates of nodes linked using connectors

· Components/Metanodes encapsulate discrete sub-workflows. Component/metanodes can be used like nodes.

· A large library of pre-build nodes and metanodes are offered for common tasks like cleaning up data, visualization, plug into Tableau and PowerBI.

· Supports all types of data.

Further reading, sources, mentions and acknowledgements.

Stefan Helfrich -- https://www.linkedin.com/in/stefanhelfrich/

KNIME – https://www.knime.com/

See the KNIME Educators Alliance and the Teaching Materials Repository.

References:

Berthold, M. R. (2019). What Does It Take to be a Successful Data Scientist? Harvard Data Science Review, 1(2)

Further reading:

For examples, additional teaching materials, sample curriculum, see “The Data Science Guide” – www.datascienceguide.org

Unless otherwise noted, the teaching materials (including workflow examples, code examples, and slides) are available under the Creative Commons Attribution 4.0 International license (CC BY 4.0).

Music

Title: Guitar House

Artist: josh pan (2020)

Source: https://www.youtube.com/watch?v=JL-LId8ZWB M

License: License CC BY 3.0

Cover Art

Title: We need You! Visual Analytics

Artist: Nuno Machado and Allen Higgins

Source: vignette_version.pptx

License: CC BY-NC-SA 4.0

Podcast License

Design Talk (dot IE) CC BY-NC-SA 4.0

The license can be viewed at https://creativecommons.org/licenses/by-nc-sa/4.0

By taking part you give permission for your voice to be recorded, for the recording to be edited, and for it to be posted and published as a podcast.

Hosted on Acast. See acast.com/privacy for more information.