Predictive text…what is it and how does it work?
Whether you’re searching for something on Google, texting a friend or drafting an email, you might notice that your device or search bar starts anticipate what you’re going to type. Whether it just offers up possible word choices, or autofills what you’re likely trying to say. You might also notice that your device gets better at this over time. This is machine learning. You’re teaching your device about your typing habits, your vocabulary and your personal voice.
Why do we have Predictive Text?
Predictive text, when used for emails or text messages, is designed to speed up your messaging process. Much like the auto-correct feature, the results can be comical. As you use your phone, tablet or computer more, the more slang and idiosyncrasies it will pick up on. So, over time it becomes more and more useful.
This isn’t new
If you were an early cellphone adapter, you may have done a lot of texting using T9 software. Because cellphones used to require you to use the numeric keypad to type letters, T9 was invented to make it faster. This would pop up possible words that you could select to help you form full or partial sentences. Now that smartphones come with QWERTY keyboards and touch screens, typing and selecting words is even faster than what T9 provided.
Can you turn it off?
Yes. On most smartphones you should be able to get into the keyboard settings and turn predictions “off.”
Predictive Search on Google
Companies like Google use predictions in their search bar for the same reason you have it as part of your phone’s keyboard – to quicken the user’s experience. Google uses an algorithm based on common searches and your own personal search history. Have you ever searched for something, then almost immediately got a prediction for your next search? Feels weird, right? What’s happening is that Google suspects you’re on the hunt for something specific and uses all it’s available data to predict your next search. As Orwellian as it seems, it can be helpful.
Search for TwinBytes…
Watch our latest video!