Twitter may influence the spread of disease

The northern hemisphere flu season will soon be here. If you get vaccinated and tweet to your followers follow you doctor's surgery - or are they the kind of people who have booked an appointment?

American scientists have extracted Twitter for tweets that short and influenza vaccination, and found that the results of the prevalence of vaccination in parallel. Now they want to know about Twitter correct posture reflects vaccination or help distribute them.

Marcel Salathe from Penn State University in University Park, Pennsylvania, 478,000 tweets referring to the flu collected in late 2009 when the vaccine for swine flu pandemic was available in the U.S.. A team of students categorized 10 percent of tweets as for, against or neutral on vaccination. Then these tweets used to design an automated screening test that the rest classified create.

Each tweet out information about the region it came from. The team found that in areas with lower vaccination rates tend to be more negative tweets about vaccination, and vice versa.

Such mapping could help to address health information campaigns, Salathe said. "If we know where people are very poorly informed, so we know what we must do better on the information."

Echo Chambers
Salathe also uses Twitter to a number of other health problems including obesity, social attitudes that may influence the distribution of the track.

Flu Tweets flowed "echo chambers", especially among people who agree with each other. Salathe now trying to find out if it's because people talk to people who think like them or think like people who speak them. He is developing new statistical tools to do this to tease Twitter data. "Preliminary analysis shows that it is very likely that the negative opinions of vaccination is contagious on online social networks," he says.

It would be bad news if the number of people who refuse to be vaccinated is rising. Twitter is a rich source of data to map these attitudes, Salathe said. Efforts to change the hearts and minds, might or might not work, "but if you do not know what your problem is that you never solve."