You are here:  Home  >  Data Education  >  BI / Data Science News, Articles, & Education  >  Current Article

Election 2012: The Semantic Recap

By   /  November 8, 2012  /  No Comments

There’s no such thing as too much post-election coverage, is there? Alright, maybe there is. But we couldn’t let things die down without at least a nod to those in our space that have delivered the semantic industry’s own take on the topic.

Here are a few you may want to review:

Twitris Election Insights:

“The Twitris system had an amazing night–while Nate Silver’s model might have received well deserved attention, Twitris gave better indications and insights and large majority of the polls,” wrote Dr. Amit Sheth, Kno.e.sis Ohio Center of Excellence in Knowledge-enabled Computing director and LexisNexis Ohio Eminent Scholar, in an email to us. The semantic social web application (first covered here) is a project of Kno.e.sis at Wright State University.


As no one could have missed, there’s been considerable attention paid by the media to whether President Obama’s performance post-hurricane had an impact in the election results. And, among the insights Twitris pointed out following Hurricane Sandy, and a couple of days prior to the election, was that “the structure of interaction network of top 100 influencers in the two topical communities [of each candidate] provides insights about increased positive cohesiveness for President Obama.”

During the election, Sheth also pointed to another trend Twitris spotted that also pointed to the final outcome: that it recorded a cross-over in sentiment for swing states Florida, Virginia, Ohio, Colorado and Florida. “If this holds, we are wrapping up the results by 11pm!” he wrote. The TV networks began projecting President Obama as the winner at 11.20 pm.

Parse.ly’s Web Wide Trends:

Speaking of Hurricane Sandy’s possible impact on the election, a company blog published right after Hurricane Sandy by predictive content analytics platform vendor Parse.ly (which we first covered here) discussed its Web Wide Trends feature — and what it told about who got what coverage in the midst of the disaster. Parse.ly is used by publishing outfits such as Reuters, U.S. News and World Report, and The Dallas Morning News to help them understand what stories are gaining steam or flagging, and why, with the help of NLP technology, metadata and visualizations.

Says the post by Parse.ly business intern Jason Bell, the feature shows that “over the days leading up to Sandy, Obama was actually trailing Romney in terms of total media volume. Thanks to Sandy, however, Obama has gotten a quantifiable press and media coverage bump.”

Bell also mentions the launch of Parse.ly News, built by the company CTO during the storm, which monitors the latest dispatches from top online media front-pages. Users can tour what’s live at The New York Times, Chicago Tribune, The Washington Post, and many more, for a scan, and click through to the publication if they want to go deeper than the headlines.

Presidential Election News & Twitter Tracker:

Semantic text analytics vendor Saplo in conjunction with BI vendor QlikTech’s QlikView “predicted the U.S. election by analysing Twitter and News data,” wrote Saplo founder and CEO Mattias Tyrberg in a Google Plus posting.

The Presidential Election News & Twitter Tracker let users review and analyze content by candidate, subject matter, category, key words and so on. It recorded 2-million plus tweets and close to 53,000 articles on the president since March, leading up to a 39 percent positive sentiment rating to the 32 percent rating challenger Mitt Romney earned based on 1.3 million-odd tweets and just under 40,000 news articles. Romney rose to a 35 percent positive sentiment score based just on November tweets and articles to Obama’s 39 percent for that month – a big change from March when the president had just a 40 percent positive sentiment score to Romney’s 49 percent (before he was the official Republican nominee).

What may be of particularly interest, though, now that the election is over, are some of the individual topic findings and what challenges that might mean for the president going forward. Taking a look at sentiment for the president about the economy, for instance, Obama records just a six percent positive rating based on tweets and articles analysis. Social security and the debt are at 5 percent, and health care and role of government stand at four percent. Foreign policy’s a bright spot, at 20 percent positive sentiment, as is Afghanistan at 32 percent.

Topsy Twitter Political Index

For a state by state look at sentiment around the president and his former challenger based on what was being shared on Twitter up through the early morning hours of election day, you can still take a look at this visualization from Twindex (which we told you about here). As of that point, its analysis of the data showed that the Twitterverse that began as overwhelmingly pro-Obama saw positive sentiments gradually shifting toward Romney, and that Montana, California, New Mexico, Illinois and Kentucky were currently the most neutral states.

Back at the main Twindex hub, as of Nov. 7, not surprisingly, Obama was up 11 on positive sentiment to a score of 85, while Romney was down 2 to 57. You know what they say, to the winner go the sentiment spoils.




About the author

Jennifer Zaino is a New York-based freelance writer specializing in business and technology journalism. She has been an executive editor at leading technology publications, including InformationWeek, where she spearheaded an award-winning news section, and Network Computing, where she helped develop online content strategies including review exclusives and analyst reports. Her freelance credentials include being a regular contributor of original content to The Semantic Web Blog; acting as a contributing writer to RFID Journal; and serving as executive editor at the Smart Architect Smart Enterprise Exchange group. Her work also has appeared in publications and on web sites including EdTech (K-12 and Higher Ed), Ingram Micro Channel Advisor, The CMO Site, and Federal Computer Week.

You might also like...

Machine Learning vs. Deep Learning

Read More →