Twitter Able to Detect Disease Trends
Lousiana - An expert in computer science at Southeastern Louisiana University showed that the trend of diseases such as pandemic influenza has the potential to be monitored much faster and cheaper through social networks like Twitter than following the traditional method.
A process called syndromic patients using health-related data collected for health officials warned of possible outbreaks of diseases such as influenza or other infectious diseases. This technique involves collecting data from hospitals, clinics and other sources.
Well, by monitoring social networks like Twitter, researchers can capture comments from people with colds that send via a status message.
"A microblogging services like Twitter are a promising source of new data to Internet-based patient because of the volume of messages, frequency and public availability," said Aron Culotta, assistant professor of computer science, as quoted by Science Daily, Monday (09/04/2010 .)
"This approach is much cheaper and faster than having thousands of data of hospitals and healthcare providers fill out a form every week," he added.
Moreover, these reports have limited time, and can not be done fast. In his research, Culotta and two student assistants to analyze more than 500 million Twitter messages during the period of eight months, from August 2009 to May 2010, collected using Twitter application programming interface (API).
By using a small number of keywords to track the level of influenza-related messages on Twitter, the team was able to estimate the level of influenza in the future.
"Once the program is running, it really is not time consuming or expensive," he said.
Southeastern Group obtained a 95 percent correlation with the national health statistics collected by the CDC. In addition, the results are comparable to figures collected by Google with its Flu Trends service, which tracks the level of influenza by analyzing the trends in terms of demand.
A process called syndromic patients using health-related data collected for health officials warned of possible outbreaks of diseases such as influenza or other infectious diseases. This technique involves collecting data from hospitals, clinics and other sources.
Well, by monitoring social networks like Twitter, researchers can capture comments from people with colds that send via a status message.
"A microblogging services like Twitter are a promising source of new data to Internet-based patient because of the volume of messages, frequency and public availability," said Aron Culotta, assistant professor of computer science, as quoted by Science Daily, Monday (09/04/2010 .)
"This approach is much cheaper and faster than having thousands of data of hospitals and healthcare providers fill out a form every week," he added.
Moreover, these reports have limited time, and can not be done fast. In his research, Culotta and two student assistants to analyze more than 500 million Twitter messages during the period of eight months, from August 2009 to May 2010, collected using Twitter application programming interface (API).
By using a small number of keywords to track the level of influenza-related messages on Twitter, the team was able to estimate the level of influenza in the future.
"Once the program is running, it really is not time consuming or expensive," he said.
Southeastern Group obtained a 95 percent correlation with the national health statistics collected by the CDC. In addition, the results are comparable to figures collected by Google with its Flu Trends service, which tracks the level of influenza by analyzing the trends in terms of demand.
Subscribe to:
Post Comments (Atom)
0 Responses to “Twitter Able to Detect Disease Trends”
Post a Comment
######silahkan anda berkomentar sesuai hati kalian,, karena setiap koment dari sahabat semua akan menambah mannfaat bagi kemajuan blog ini di masa mendatang########