Google’s Flu Trends map is almost solid red, indicating intense flu activity across the entire United States in early January, 2013. How does a search engine gather data on flu symptoms? It’s a process called data mining, and they’ve used it before – namely during the 2009 Swine Flu pandemic.
Flu Trends: Map Shows Disease Spread
During an epidemic, information on how and where the disease is spreading is valuable to health providers and potential victims alike. Predicting where the disease is breaking out helps to ensure availability of treatment and medicine for patients, and tells us when to stay home rather than going to a crowded public area. Google developed a system in 2009, in response to the H1N1 pandemic of 2009, to analyze search queries for certain flu-related terms. They use the terms to determine whether or not the swine flu has spread to a certain area through the use of data mining, and natural language processing.
Flu Symptoms Search: Data Mining and Natural Language Processing
Data mining – most of us have heard the term before, but usually in the context of marketing. For example, social media sites may use a computer to sift through your activities online to provide the most customized advertising to you as a consumer. When we sift through large sets of data to pull out valuable information, (information which is typically not tied to a specific user or consumer,) that is called data mining.
In the case of flu symptom maps, Google analyzes search queries with natural language processing (comparing text or speech to derive meaning). The search giant uses natural language processing in day-to-day business for a number of purposes, including behavioral advertising, in which advertisements are directed toward individuals based on their search history. The search engine can also use the search information gathered for this purpose for other, less profitable, purposes, however – in this case, the Google Flu Trends map.
By scanning search queries for common flu symptoms, Google can determine whether there are increasing numbers of searches that may be due to actual flu symptoms. Associating those queries with the IP address of the source computer allows Google to associate the queries with a geographical location. The combination of these two types of information processing shows increasing or decreasing flu activity in different areas around the world. Right now, unfortunately, the Google Flu Map is showing intense activity for the majority of the United States.
Saving Lives with AI Applications
While Google’s information is incomplete and imperfect, this use of artificial intelligence technology may save lives as the 2013 flu spreads. Although Tamiflu and vaccine shortages are complicating health care providers’ efforts to combat and prevent flu symptoms, preparation is still key in responding to any health crisis.
Google. Flu Trends. (2013). Accessed January 11, 2013.
Decoding Science. One article at a time.