(Annette Moser-Wellman)

Imagine for a moment that you didn't search for news, but news searched for you. The news knew your mood. What you wanted and needed to know showed up on your television, your netbook, your computer and, especially, your mobile device. If you had trouble getting out of bed, you might get an inspirational story on the radio. If you were interested in local politics, you would be served a story on your handheld about the debate on affordable housing. About 4 in the afternoon, your computer would suggest a dinner idea and, when you came home from a hard day, the television would provide three options for a movie on-demand. The news delivery would match your behavior and your attitude.

In our latest blogs, we've been interviewing academics researching the frontiers of science and technology: men and women who have projections about the future of media. Sensors are one of the most intriguing areas of development that promise application within the news media. Sensor data is available through a plethora of electronic devices including cell phones and a variety of other mobile devices. Scientists are asking "What attitudes and behaviors could we predict from the analysis of sounds, movements, and location over time?"

Tanzeem Choudhury, an Assistant Professor in Computer Science at Dartmouth College, is one of those scientists. She researches how building machine-learning techniques can help people communicate and understand each other better. She uses sensor data such as microphones, accelerometers and GPS to determine what people are doing at any point in time. "You can tell whether someone is running to catch a bus, whether someone is having a conversation in a restaurant with friends or sitting quietly at a café. You can determine the entire range of behavior they are engaged in."

One of the focus areas of Choudhury's research is understanding the stylistic content of conversations. She looks at the data from a group of people and can determine which people are more central in a social network and those who are on the periphery. People who are more prominent in a network do not adapt their speaking rate, their turn-taking, the loudness of their voice as much as those on the periphery who adapt to the central individual. Choudhury can predict those who are the influence-makers in the social network from the voice - even without the words.

At first this might seem very Big Brother. But Choudhury believes individuals will see how behavioral monitoring can improve their lives. They will become enthusiastic and willing participants and opt into such types of tracking. She speaks about the potential for health care and how individuals could get help with managing their health and could build a better quality of life.

"If you know what kinds of environmental and behavioral factors lead to long-term health problems, you can keep track of that and give information back to the user that can lead to positive change. For example, people in the United States don't get enough physical exercise. But if you had a way of getting information to them about how much they actually took the stairs or how much time they spent in a chair, they'd have more information and control about what they could change."

From analyzing conversation and movement data, it is also possible to predict mood. When someone walks more slowly, talks more slowly, there is potential to provide the right feedback at the right time to deliver information they might need.

"I see in the future a broad set of applications that help you communicate at a better level," she says. "Now we share a piece of news story in a social network because we think it might be interesting to someone else. And how do we know it will be interesting? Because we know the person well enough – their tastes, what they do. Computers, cell phones and modern technology will have that same insight. They will be able to predict the usual, but also to predict what might be unusual and really interesting to the other."

Eventually, Choudhury sees the ability to gauge the mood of an entire city or group of individuals.

"If you think about what we are doing in terms of behavior analysis, we are assessing a person. The information can be used to characterize activities, personalities, and emotions," she says. "We can map the current mood of individuals and groups in an area. Potentially you can cater news to a city in Italy that was struck by an earthquake or a city like Rio after Brazil won the World Cup. Knowing the emotional state of a group of individuals determines what kind of information someone might be receptive to."

So expect news organizations to be able to get a lot smarter about whom they serve, what to serve and when to serve news. One of the complaints about Web sites, newspapers and especially television, is that it takes so much time to search for what's interesting. Technologies like those Choudhury is tackling could eventually level those concerns. News companies that search the news for you and give you what you didn't even know that you needed will be the winners in the future.


Annette Moser-Wellman is President of Firemark, Inc., an innovation consultancy, and author of Six Competencies of the Next Generation News Organization and Running While The Earth Shakes: Creating An Innovation Strategy To Win In The Digital Age, both published by the Media Management Center.

This TechScout article is part of a series of Moser-Wellman interviews commissioned by the Media Management Center to explore opportunities and insights at the intersection of technology and the news media. Click here to view other articles in the TechScout series.

What do you think? Please share your thoughts, experiences and reactions by clicking on the comment button below.