[AI] Experts more influential than celebrities on Twitter
gopalakrishnan_vip at yahoo.co.in
Fri Sep 10 22:39:05 EDT 2010
Experts more influential than celebrities on Twitter
Washington, September 10, 2010
Just because you have an army of followers on micro-blogging site Twitter doesn’t mean you’re among the most influential people in the Twitterverse, say experts.
While celebrities gain huge followings in the Twitterverse, the top influencers on the hot topics of the day are likely to be people with much lower profiles but experts in their own rights.
If you really want to know the most influential people tweeting on the hot topics of the day, go to www.pulseofthetweeters.com. The website went live in May and has been tracking the top trending topics from Twitter in real-time ever since.
The website was created in the lab of Alok Choudhary and John G. Searle, professor and head of electrical engineering and computer science at the Northwestern University, US, and grew out of the Ph.D. thesis of Ramanathan Narayanan.
“The question we’re really asking is: whose opinions are most interesting and influential on any given topic?” Narayanan said, according to a Northwestern release.
“If someone from BP is tweeting about the oil spill, for example, his opinions are likely to carry much more weight and be of much greater interest than those of Ashton Kutcher, who has a legion of followers,” Narayanan said.
“There are about 50 million tweets produced every day, but most of us only read 10 or 20 tweets in one sitting,” Narayanan said.
“So, which tweets should you read? Which tweets are being read by media experts on any given subject, such as politics, law, fashion, food? We provide that information for users,” Narayanan said.
The website uses a specialised algorithm to rank the most influential people tweeting on trending topics.
The algorithm for the website combines dynamic data mining, (process of extracting patterns from data), sentiment analysis and network analysis in real-time.
Besides identifying the most influential tweeters, the algorithm can tell you whether their tweets are positive, negative or neutral. It also offers related topics to explore.
“Discovering patterns, opinions and sentiments from massive number of tweets is challenging in itself, but discovering influencers and leaders for specific topics is a major technological advance in data mining,” said Choudhary, also a professor at Northwestern’s Kellogg School of Management.
“The good thing about our system is it’s completely automatic, and it needs minimal human supervision,” Narayanan said.
More information about the AccessIndia