What are people looking for on Google and Wikipedia?

Exercise: What are people looking for on Google and Wikipedia

Google logs everything that people look for. The surprising news is that Google makes some of this data available to us!

Here, we are going to use a website called Google Trends to have a look at this data.

You can use an existing Google account if you have one, or signup for one here (free).

Load up your preferred browser. Type in the URL trends.google.com. This website allows you see what people are looking for on Google. You can type in search terms to see how frequently people have looked for them.

The World Cup is an event which fascinates a large number of people worldwide – so let’s try it with the term “World Cup”. Don’t forget to hit the return key or to click on the search button. What do you find?

Now let’s try it with the term “summer”. What do you find? In which months do people search most for information about the summer?

Below the search field, it says “Worldwide”. This tells you that you are looking at data generated by users around the world. Click on “Worldwide” and select “United Kingdom” instead. Do you see any differences? What happens if you pick “South Africa” or “Australia” instead?

Now let’s find out which countries like which types of sport. Type in some names of sports in the search field separated by commas: for example, “tennis, cricket, football, soccer, baseball”. Try different countries such as the United States, UK, India, and Germany. What do you see? What language do people speak in these countries? Might this affect how you can interpret these patterns?

Finally, let’s see if we can find any differences in how women and men look for presents for their partners.

Set Google Trends to give you Worldwide data again. Type in the two search terms “gift for boyfriend, gift for girlfriend”. What do you find?

Google Trend: gift for boyfriend, gift for girlfriend

Now compare your results with a second pair of search terms: “gift for husband, gift for wife”. What do you find?

Google Trend: gift for husband, gift for wife

If you would like to look at the last year in more detail, you can change the setting “2004 – present” to “Past 12 months”. When are there more searches for “gift for husband”? When are there more searches for “gift for wife”?

In order to download data from Google Trends, you need a Google account, in particular for exercises in subsequent weeks. If you have one, log in now. If not, make one and then log in.

If you are logged in, you will see three little dots in the top right hand corner underneath your email address. Click on this sign, and then click on “Download as CSV”. You will then be able to download a file you can open in Microsoft Excel, or another spreadsheet program. We will look at how to work with this data in later exercises.

Download data as CSV from Google trends

For those of you who are unable to access the Google Trends website you can download the results, as a CSV file, here.

Not only Google makes data on usage of its service available, but other Internet based services too. Another example is Wikipedia. Most people know that Wikipedia makes articles on a range of subjects available for free. However, there is also a service which will give you free access to data on how many people have been looking at these articles: http://stats.grok.se/.

Go to this website and type in the article name “Easter”. You can increase the default time period of 30 days by clicking on “90” days. What do you see?

Try typing in “European Central Bank” instead, and increasing the time period to 90 days. Where do the spikes come from? Stories in the Guardian might help you work this out: http://www.theguardian.com/business/european-central-bank

Wikipedia - European Central Bank

Congratulations! You’ve made your first steps in becoming a data scientist and know how to find out what people are looking for on Google and Wikipedia.

What other interesting examples of patterns in Google and Wikipedia data can you find?

Let us know in the comments below - or start a conversation on Twitter, remembering to use the hashtag #FLbigdata, or by tweeting us at the course Twitter account @BigDataMOOC or at @t_preis or @suzymoat.

Thanks to Hal Varian for the Google Trends ‘gifts’ example.

Share this article:

This article is from the free online course:

Big Data: Measuring and Predicting Human Behaviour

The University of Warwick

Get a taste of this course

Find out what this course is like by previewing some of the course steps before you join: