Are you interested in the future of user interfaces for business intelligence?
Tomorrow (Wednesday, November 4th 2009) there is a CASCON workshop on "User Interfaces for Visual Analysis and Monitoring in Business Intelligence". It is co-organized by Margaret-Anne Storey, Christoph Treude and myself. The workshop starts at 1pm in Markham Ballroom C. This is the agenda:
1:00 pm Introduction (Margaret-Anne Storey, UVic)
1:15 pm Charting and Visualization at Cognos Software (Stephan Jou, IBM)
1:45 pm Mapping the BI Visualization Landscape (Mike McAllister, SAP)
2:15 pm Visualization Construction by Non-Experts (Lars Grammel, UVic)
2:45 pm Visualizing higher-dimensional Pareto fronts for complex decision making (Derek Rayside, MIT)
3:15 pm Coffee break
3:30 pm Dashboards in IBM's Jazz: BI for Software Development (Christoph Treude, UVic)
4:00 pm Business Intelligence on Mobile Devices (Stephan Jou, IBM Cognos)
4:30 pm Wrap-up (Margaret-Anne Storey, UVic)
We hope to see you tomorrow!
Subscribe to:
Post Comments (Atom)
33 comments:
Stephan just presented 13 interactions that tools support for visualizations. These all seemed to be human-to-chart interactions. What are the key human-to-human collaborative interactions that BI tools do or should support?
Stephan talked about the chart advisor for Microsoft Excel and that it was good, but limited -- in what ways is it limited? How can semantics be leveraged to improve the charts?
From Mike McAllister's talk - he mentioned work on visualization of unstructured data at UBC (through text summarizations), is this the work he was referring to:
http://people.cs.ubc.ca/~carenini/PAPERS/iui06.pdf
The main dimensions of the BI Visualization landscape:
- skill level of the user
- interactivity
- delivery medium
Which other dimensions can we think of?
@Christof Is the format / type of underlying data a dimension? Also, he had "number of users" underlying all the dimensions
Delivery medium and user skill are perhaps multi-dimensional variables?
If our goal is to integrate our visualizations in other tools and media -- i.e. "free" them from the tyranny of the proprietary creation tool -- do we have standards for this? (can I import/export views between IBM and SAP tools)?
OLAP: http://en.wikipedia.org/wiki/Online_analytical_processing
Is decision support vs. exploration another dimension?
@Kelly: I would think so, but I guess we're assuming lots of somewhat structured data here.
Discussion -- how to support the discovery of an emergent schema in unstructured data?
Re Lars' presentation -- could the same wizard of oz approach be used for determining requirements for Human-Human collaborations during visualization creation?
Re Lars' talk, discussion on how do we support "messing with the data"
How to also support exploration of simpler views of the raw data?
Derek is interested in collaborations when he moves to Canada (Waterloo) in the Spring as a faculty member!
This is Derek's CASCON paper: http://sdg.csail.mit.edu/pubs/2009/cascon09-moolloy-ui.pdf
He's exploring future work in his talk at this workshop.
Petal diagrams: http://www3.interscience.wiley.com/journal/5728/abstract?CRETRY=1&SRETRY=0
Reduce s -- would clustering them or grouping them work? Certainly not always ...
meant reduce |s| ...
Reminds me of this slashdot article from last year: http://science.slashdot.org/article.pl?sid=08/08/24/0240230.
How to integrate push and pull mechanisms in dashboards/feeds? (Christoph's talk)
Stephan's talk -- how to take complicated charts and put on constrained mobile devices?
Preliminary version of key issues from our workshop:
- Business intelligence for users of varied skill levels – what can we assume, if anything?
- Target audience: chart authors, chart users, chart integrators,…
How to provide effective suggestions or recommendations?
- How to extract semantics? How to incorporate user training in the process?
- Human-human collaboration in business intelligence, what kind of tool support is needed?
- Balancing aesthetics vs. effectiveness?
- How to introduce visualizations for more complex tasks (e.g. pareto fronts)?
- Support for multiple media (e.g. small screen displays)?
- Use of standards, will it be straightforward as business intelligence becomes more advanced and more dynamic?
- How to integrate push and pull information retrieval mechanisms? (need to be careful how notifications are done, need flexibility)
- How to address challenges in the wireless space (security, routing, UI support, different standards, hardware constraints)
- Serendipity: what is the right level of exploration vs. decision support?
- ???
More Powerful mobile development (Stephan's talk): The audience is blogging about him during his talk...
Additional key issues:
How to extract "best practices"
A lot of users don't know that they are consuming BI --> we need to serve these users as well!
cultural aspects: how people interpret information
data integration
Thank you for a great workshop!
Interesting! I've created a tag in delicious to track the URLs I've started collecting 'CASBUIUI' at http://delicious.com/seemsartless/CASBIUI
Additional simple semantics that have a big impact on visualizations: ordinal (vs nominal), location, currency, time. And those are the simple ones! We have so much more in a typical BI metadata store that we can leverage, but currently are not.
Having seen what millions of rows of raw data look like, I suspect there is opportunity for guided exploration and visualization there, without introducing excessive information hiding or inhibiting unfettered/unbiased exploration.
Thanks for the great participation and comments yesterday -- we will integrate our full set of issues and post them shortly!
Thanks for the participation in our workshop! I have posted the slides of the talks (see blog post above).
Post a Comment