WebProNews

Tag: Google Correlate

  • Google Correlate Expands to 50 Countries

    Back in May, Google launched Google Correlate, which is sort of like Google Trends in reverse. It looks at search trends, and attempts to apply them to real-world situations.

    Google described it as an experimental tool enabling the user to find queries with a similar pattern to a target data series.

    Users can upload data sets (by state or time series, and Google Correlate will compute the “Pearson Correlation Coefficient” between your time series and the frequency time series for every query in its database.

    Google Correlate

    Previously only available in the U.S., Google announced today that the service is being extended into 49 new countries.

    “Since our initial launch, we’ve graduated to Google Trends and we’ve seen a number of great applications of Correlate in several domains, including economics (consumer spending, unemployment rate and housing inventory), sociology and meteorology,” says Google software engineer Matt Mohebbi. “The correspondence of gas prices and search activity for fuel efficient cars was even briefly discussed in a Fox News presidential debate and NPR recently covered correlations related to political commentators.”

    “Health has always been an area of particular interest to our team (Matt Mohebbi, Julia Kodysh, Rob Schonberger and Dan Vanderkam),” says Mohebbi. “Correlate was inspired by Google Flu Trends and many of us worked on both systems. So we were very excited when the BioSense division at the CDC published a page which shows correlations between some of their national trends in patient diagnosis activity and Google search activity. With just three years of weekly data, relevant search terms are surfaced. For example, the time series for bloody nose surfaces ‘bloody snot’ and ‘blood in snot’.”

    “While these terms shouldn’t come as a surprise, there are others which are more interesting, including searches related to static electricity, dry skin, and red cheeks,” says Mohebbi. “Of course, correlation is not causation but we hope that Correlate can be used as a method for researchers to generate new hypotheses with their data.”

    For more on Google Correlate, check out this FAQ page.

  • Google Correlate Launched – Google Trends in Reverse

    Google just launched a new Google Labs product called Google Correlate, which looks at search trends, and attempts to apply them to real-world situations. The official description for Google Correlate is as follows:

    Google Correlate is an experimental new tool on Google Labs which enables you to find queries with a similar pattern to a target data series. The target can either be a real-world trend that you provide (e.g., a data set of event counts over time) or a query that you enter.

    It uses search activity data to find queries with a similar pattern to a target data series – the results of which can be viewed on the Google Correlate site (or downloaded as a CSV file).

    “Google Correlate is like Google Trends in reverse,” Google says on an FAQ page. “With Google Trends, you type in a query and get back a data series of activity (over time or in each US state). With Google Correlate, you enter a data series (the target) and get back a list of queries whose data series follows a similar pattern.”

    Google Correlate

    Users can upload their own data sets. When you upload one, (by US State or Time Series), Gogole Correlate will compute the “Pearson Correlation Coefficient” between your time series and the frequency time series for every query in its database.

    Google provides a tutorial on how to use it here. There’s also a whitepaper.

    Google suggests the time series data can be used to find things like what search terms are more popular in the winter, more likely to be issued in 2005, match the pattern of actual flu activity, etc. The state data can be used for things like what terms correlate with the state’s latitude, the annual rainfall in the state, being in New England, etc.

    Interestingly the Labs experiment has its own Labs section, which so far only consists of one thing: search by drawing. It’s pretty cool. You can simply draw a pattern on the graph, and it will give you web search activity that closely matches the pattern you drew. I drew some random pattern, and Google found that it closely resembled US web activity for media player 10 codecs, for example.

    Google Correlate Drawing

    Google says the data for Google Correlate is available from January 2003 to the present, with data being updated on a weekly basis.