November 2012, Vol. 13, No. 11

• Trending on Twitter, Five Hours from Now
• In Need of a New Systems Approach to Save the World’s Fish
• Social Media Provide Early Alerts for Manufacturing Defects
• Video Highlights from WorldFuture 2012
• What's Hot in THE FUTURIST Magazine

Trending on Twitter, Five Hours from Now

MIT Researchers Devavrat Shah and Stanislav Nikolov have created a machine learning algorithm that can predict what topics will trend on Twitter as many as five hours in advance, with 95% accuracy.

"This is a very simplistic model. Now, based on the data, you try to train for when the jump happens, and how much of a jump happens," says Shah.

The algorithm determines the probability of a topic becoming a trending topic based on how closely the topic in question resembles the growth pattern of previous topics that went trending. As the amount of training data grows—that is, the number of raw tweets from Twitter that reveal patterns--the algorithm will improve, says Shah.

Speeding up the identification of breaking news and hot topics could benefit news organizations, users, and advertisers, the researchers believe.

Source: MIT

Devavrat Shah will speak about this research at the MIT Museum on Friday, Nov. 9, from 12:10 to 12:50 p.m. Learn more here.

In Need of a New Systems Approach to Save the World’s Fish

Fisheries management as we know it is failing the world’s fish, says Ellen K. Pikitch, executive director of the Institute for Ocean Conservation Science and professor at Stony Brook University. She warns that fisheries will have to become much more precautionary, or else many of the world’s fish species will die out.

In an article for the October 26 edition of the journal Science, Pikitch and co-authors point out that fish populations worldwide are all below the levels that standard fisheries management recommends. Even worse, many species are exhibiting a continuing trajectory of decline.

Traditional fisheries management makes the mistake, the study contends, of focusing on only one given species at a time and not looking at the whole ecosystem or at the other fish species and wildlife whose numbers may be declining, as well.

Pikitch serves on the Lenfest Forage Fish Task Force, which proposes replacing this "single-species" approach with an " ecosystem-based" approach. This would limit fishing strictly to the places and quantities that are safe and that pose minimal risk of damaging the underwater ecosystems, based on what the data of the ecosystems and fisheries indicates.

Source: Stony Brook University

Social Media Provide Early Alerts for Manufacturing Defects

If something goes wrong with your car, you’ll probably text your family first, Facebook your friends, Instagram a snapshot of your disaster, and tweet a general complaint to the rest of the world. Then you’ll get the car fixed. Probably the last thing you would do is go to the manufacturer’s Web site and fill out a feedback form.

Social media represent a largely untapped source of information that could be extremely valuable to manufacturers that need to stay on top of safety and reliability problems. The challenge is to sift useful data out of the mountains of unrelated information that consumers share on message boards, blogs, Facebook, Twitter, and the like.

At Virginia Tech, business information scholars Alan Abrams and Weiguo Fan are developing a sort of linguistic smoke detector to identify potentially useful information about specific automobiles that may be contained in vast amounts of " dynamic and unstructured" social media content.

Beginning with online discussion forums for owners of Honda, Toyota, and Chevrolet vehicles, the researchers employed car experts to manually sort and tag posts that contained information about defects. The "automotive smoke words" enabled the researchers to devise decision support systems (such as automated Web crawlers) to help manufacturers discover defects.

The researchers plan to expand their analysis to Twitter and Facebook postings. Says Abrams, "With the volume of social media posts expanding rapidly, we expect that the need for automated business intelligence tools for the exploration of this vast and valuable data set will continue to grow."

Source: Virginia Tech

WorldFuture 2013: Exploring the Next Horizon

The Annual Conference of the World Future Society: July 19-21, 2013 at the Hilton Chicago Hotel, Chicago, Illinois.

Register Today and Save $250!

The World Future Society's annual conference, WorldFuture 2013: Exploring the Next Horizon, will give you the opportunity to learn from others in many different fields, and to explore actions affecting our futures in as yet unimagined ways.

The conference will feature nearly 100 leading futurists offering more than 60 sessions, workshops, and special events over the course of two and a half days. And for those who want to take a deeper dive, into key studies of interest, the preconference Master Classes allow for an in-depth look in a small group setting.

New for 2013: 22nd Century Lecture Series

Special hour-long sessions each day will focus on one of the six major themes of the conference, offering expert insights on issues, trends, forecasts, scenarios, and wild cards in Earth, Humanity, Commerce, Governance, Sci/Tech, and Futuring.

Video Highlights from WorldFuture 2012

Age Isn't a Number (Video)

Adriane Berg, author of How Not to Go Broke at 102, explains why insurers don’t care how old you are – and neither should you. Excerpt from WorldFuture 2012; video by Hunter Molnar Stanton for the World Future Society. Watch the video here.

What Is A Futurist? Interview Nine of Them in Nine Minutes

By Patrick Tucker

What is a futurist? Every self-described futurist you ask will likely give you a different answer. A more interesting question is what do these people say about the future? Brian Bethune from Maclean's magazine recently put that query to a grab bag of inventors, technologists, geneticists, business consultants, and writers he encountered at WorldFuture 2012, the Society's recently concluded conference in Toronto, Canada. Read more.