Google using Machine Learning to predict flight delays
San Francisco: Google has introduced a new feature in its "Google Flights" service that will help users predict flight delays and find the cheapest fare by using machine learning capabilities.
"Google Flights" -- an online flight booking search service which facilitates the purchase of airline tickets through third party suppliers -- will now surface information for flights on American, Delta and United Airlines, Anket Mathur, Flights Product Manager at Google, said in a blog late on Wednesday.
"New fare types like 'Basic Economy' are great for people travelling on a budget but it can be confusing to understand whether important options -- like overhead bin space, ability to select your seat and baggage fees -- are included in the fare," Mathur wrote.
Using historic flight status data, Google's machine learning algorithms can predict some delays even when this information is not available from airlines yet.
"The delays are only flagged when we are at least 80 per cent confident in the prediction. We still recommend getting to the airport with enough time to spare, but hope this information can manage expectations and prevent surprises," Mathur added.
Google will also provide reasons for the delays, like weather or an aircraft arriving late.