Previous Blog Entry Next Blog Entry

This session was presented by Emily Wheeler and Samara Omundson.

In 2009 the digital universe grew immensely.  If you picture a stack of DVDs reaching to the moon and back, that’s how much data growth we had.  By 2020 it is estimated that the 44 times the size it was in 2009.  We are drowning in data.  How can we make sense of it?  Apply structure to large quantities of data to help make sense of it.  Information professionals can lead the way through these piles of data, even if we’re not statistics junkies or graphic designers.  We know how to sift through vast quantities of data and pull out those few salient data points.  Data conveys a clear message, cuts through the chaos, and helps to engage and inform stakeholders.

One strategy for information visualization is using topic clusters.  As an example, they searched for “Bieber Fever” with a general search engine.  They displayed results in a hierarchical format using a single PowerPoint slide.  Another visualization was a branching choice – almost a spider web of information circulating out from a central point.  Another strategy is using time series visualizations.  These can be line graphics or bar graphs of a particular data point’s change over time.  You highlight an intersection in searches using Search Associations.  You can do this in spreadsheet tools, but they used a great tool called TouchGraph to create some really nice relational graphics.

How do you handle text analysis differently?  Keyword frequency is very useful to identify repeated keywords—a simple word count provides this data point. Creating a bar graph quickly with the number of mentions of various words can show the relative importance or permeation of various words and ideas.  You can use Tagzedo to create good keyword clouds.  By adding word association to simple keyword frequency you can see relationships between words and concepts.  Using different colors, sizes, and boldness of visualization elements can communicate the relative importance quickly.  Structural data, like keyword associations, focus on word order.  This helps you drill down into the context of a given word.  They used IBM’s ManyEyes tool to create a really nice looking structural chart.  Looking at social media data, Twitter and Facebook activity and followers, can tell a really compelling story about how social interaction and popularity relate to frequency of posting, where you post, etc.  They built a few visualizations in Adobe Illustrator.  Visuals tell a story, they show patterns.  They touched on Infographics quickly.  An infographic is a visual representation of information, but most are designed to tell a visual story of pretty complex data.  You see these in large media outlets in articles and in lead stories.  It is really easy to transform data – try it out!

Tips and tricks: Know your message, stay simple, and experiment with data visualizations whenever you can.

Leave a Reply

LiB's simple ground rules for comments:

  1. No spam, personal attacks, or rude or intolerant comments.
  2. Comments need to actually relate to the blog post topic.