Artificial Intelligence (AI) is being considered as a fundamental piece of tech when it comes to making the machines ‘smart’. Artificial Intelligence helps the machine imitate human intelligence and helps in their thinking and decision making process. With the help of AI, machines can now learn through experiences by themselves with little to no involvement from humans.
Video Credits: Aalto University, YouTube
Scientist and researchers all over the world are trying to understand reasons behind certain human actions and are making machines capable of imitating it. Whenever we use a touch screen phone, one usually relies in their sense of sight while moving fingers on the displayed keyboard. Since one can’t feel the keys we must also see the word typed out on the screen to rectify any errors. There are two tasks being performed at the same time.
In order to understand the process behind how people type on the touchscreen, a team of researchers at the Aalto University and the Finnish Center for Artificial Intelligence (FCAI) have now come up with an AI model which is capable of predicting how the users move their eyes and fingers while typing.
Details about this study are published in a paper titled, “Touchscreen Typing As Optimal Supervisory Control”, and will be presented at the ACM CHI conference on 12 May 2021.
This project was headed by Dr Jussi Jokinen. Speaking about the study, they said, “Previously, touchscreen typing has been understood mainly from the perspective of how our fingers move. AI-based methods have helped shed new light on these movements: What we’ve discovered is the importance of deciding when and where to look.”
They also added that, “Now, we can make much better predictions on how people type on their phones or tablets.”
The AI model developed is capable of imitating a human users typing style on any keyboard design. Just like any other human user, this system makes errors and is also capable of detecting it and corrects itself. Along with this, the simulation is capable of predicting how people change their style of writing when they start using a new auto correction system or a keyboard.
One of the major benefits of this system is that it can be used to develop typing aids and interfaces specially for people suffering from motor impairments. This system makes use of one of the most widely used machine learning techniques, reinforcement learning. Though reinforcement learning is typically used to make robots learn through trial and error method, this system is also makes use of this by detection of errors and rectifying them.
Providing a brief explanation about the system, Dr Jussi Jokinen said, “Now that we have a realistic simulation of how humans type on touchscreens, it should be a lot easier to optimize keyboard designs for better typing—meaning fewer errors, faster typing, and, most importantly for me, less frustration.”
The model’s predictions were tested by comparing it with human typing and it was found to be accurate. For the future, the team wants to develop slow and fast typing techniques. These are expected to be useful for those who want to improve their typing.
Dr Jussi Jokinen said, “We gave the model the same abilities and bounds that we, as humans, have. When we asked it to type efficiently, it figured out how to best use these abilities. The end result is very similar to how humans type, without having to teach the model with human data.”