IMAODBC 2016: And the winner is…

The Best Presentation Award of the International Marketing and Output Database Conference IMAODBC 2016 in Gozd Martuljek, Slovenia goes to Susanne Hagenkort-Rieger and her team from DESTATIS (Statistisches Bundesamt, Germany).

In her presentation Susanne highlighted the importance of web search statistics  and why intuition when emphasizing selected statistical data is often not sufficient. To achieve relevance and accessibility of most popular statistical data we should not ignore what the web search data say.

a12fd9e2e59120fadd5cfb91863759d1

Presentation is available at the IMAODBC 2016 website…

A few facts about IMAODBC 2016 as presented in the second best presentation by Corey Jenkins, USDA – Foreign Agricultural Service, U.S.A.:

IMAODBC-2016-ig.jpg

 

 

Wolfram’s CDF

Two years ago Wolfram launched Wolfram|Alpha, a search engine (‘computational knowledge engine’) which does more than find objects or link objects in the sense of linked data:  An engine computing answers from a huge amount of (partially manually) curated data and milllions of algorithms used  in Wolfram’s software Mathematica.

(not the freshest data by the way ;-))

‘Oh, and by the way, these days the majority of queries to Wolfram|Alpha give zero hits in a search engine; they don’t ever appear literally on the web. So the only way to get an answer is to actually compute it.’
So the words of Stephen Wolfram in his keynote speech at Wofram Summit 2010. This speech gives an extensive insight into Wolfram’s philosophy and objectives, fascinating!.

CDF

And now, some days ago, Wolfram launched a new data format – CDF: Computable Document Format. From the announcement: ‘CDF is a new standard that’s as everyday as a document, but as interactive as an app. It empowers readers to drive content and generate results live for a deeper understanding. And authoring interactivity is easy enough for teachers, journalists, analysts, managers, or researchers to add to reports, presentations, blogs, infographics, articles, and textbooks.’

A viewer (100mb download) is needed to read the CDFs and Mathemathica (Wolfram’s Software) is needed to edit a CDF.

Take this example: Age Distribution in the World as a CDF document. It’s 78kb to download and presents this ternary graph:

This ‘ternary diagram is a graph that shows the proportion of three variables as a position in an equilateral triangle. The three variables have a constant sum (in this case to unity). This particular diagram shows the population proportion of children (0<=age<15), adults (15<=age<65), and elderly (65<=age) for different countries. The proportions have been color-coded to facilitate interpretation. …  You can choose a continent or the whole world.’  (manually or with the autorun function).

More examples –> here.

The code underlying this presentation is Mathematica (first lines only):

This code of a CDF  is generated by Mathematica:

CDF and dissemination of official statistics

Till now I haven’t seen a lot of Mathematica usage and visualisations in official statistic’s dissemination. CDF could be an interesting tool for interactive offline publications in this field.

But it’s not an open standard and as an alternative there exists PDF which allows to embed interactive elements (flash, RIA), too (rarely seen).

From Cinema Madrid to Unemployment Spain

Some days ago in a post I mentioned how Google and others go semantic and provide in their search results not only information about information (means: links to web pages) but information itself.  So i.e. the cinema showtimes.

And Googles does even more. Google search directly provides statistical information.

Unemployment Spain gives this:

In context: (Hint: use Google without country redirect, this is: google.com/ncr)

And the source is Eurostat via Google Public data Explorer:

So why go to Eurostat or another statistical site ;-).

Pivot – a new way to look at data

Last month, Microsoft Live Labs released a new experiment called Pivot, which allows interaction with large data sets through an easy-to-use interface. Users can work with vast amounts of data, shifting from broad to narrow queries and back again. Watch Gary Flake’s TED2010 talk to see it in action.

Pivot can be downloaded from www.getpivot.com/download, but requires Windows 7 or Vista and IE8. Being a Mac / Windows XP user, I’m keen to hear from others who have managed to install and experiment with this. Any applications for official statistics?

Statistics Sweden social media contribution to Eurostat “Sponsorship on Communication”

http://swedeneurostat.blogspot.com/

The blog  is an example on how to extract and visualise Eurostat statistics. It is a Statistics Sweden prototype within the Eurostat project called “Sponsorship on Communication”. The project is initiated by the Spanish NSI during their present EU chairmanship. All member countries have been invited to contribute.

The idea from Statistics Sweden is to focus on Sweden in the European statistical context in a blog based on publishing from Eurostat, e.g. Statistics in Focus and other European data . Information and data connects to social media functionality. It can be RSS flows based on regular Eurostat and Statistics Sweden publishing and on blog and Twitter searches. Any other country within EU can apply the approach to their own country from this framework.

The search functions are from Google. The information flow in the blog list is constructed from Google Blogger widgets. New ones can be added and obsolete ones can  easily be replaced. They automatically update the flow of  information once they are defined.

In the module “Latest Search Results about Eurostat and Sweden”  pre-defined search terms are presenting results directed to different  media like images, videos, news, blogs, updates, books and discussions. They can also be categorised in time, geography and related to different search terms.

By enter key words at  “Search this blog” extended alternatives are presented to widen the scope of search to sites linked from the blog, the blog list, Statistics Sweden and Eurostat publishing and finally the whole web. So universal search is combined with more narrowly defined search destinations depending on how the blog is organised.

There is also news roll displaying information from different news media continuously through a widget. Also this information is updated automatically and search words can be changed or extended.

The scope and functionality will be tested and developed. It has been presented to journalists and media so far. We have recently included also an example of interactive publishing including statistical story telling through Vislets, developed by Professor Mikael Jern, NCVA, Linköping University, Sweden. Click on picture below for only this entry or on Sweden in Eurostat for the entire blog.