Pascal

Pascal

Inside Google Search

Google Search

Louis XIV passion for learning lead to exploring new worlds. Searching for knowledge is in our DNA long before Louis XIV and is still relevant today.

Google Search has changed our way of finding answers. We get what we're looking for in a fraction of second compared to the time needed to look for information in encyclopedia for example.

But sometimes the results we're looking for are not relevant. Google has built algorithms to rank web pages so that they correspond to our searches.

An algorithm can be defined as a set of mathematical instructions fed to a computer to achieve certain tasks. It's like a recipe. Google search algorithms are well guarded secrets, at the same level as the Coca Cola recipe. One of the reasons is to prevent hackers in corrupting the search, making spam pages appear at the top of the results.

Then, there is predictive typing (or auto suggestion) which allows you to autocomplete your search. Sometimes the search gives inappropriate (or which can be interpreted as offensive) suggestions. The latter can be reported to Google to improve the prediction algorithms.

Google is also experimenting lots of things to make it more natural to ask Google Search find what you're looking for. Experimentations on Natural Language Processing using deep learning (BERT/DeepRank) are carried out.

To summarize, we can say Google Search is like a guide that will open the door on the vast universe of the Web and help you find what you're looking for.

Additional resources

A great story telling video on Google Search can be found here.

A guideline on search quality rating can be found here