article A company is using a machine learning technique to develop a “software in your system” system that can automatically detect if a user is sick or has a problem and then provide alerts that can be taken care of automatically.
Key points:The technology, called ‘discovery’, has been used in other products, including home automation systemsThe technology can detect whether a user has a condition such as a sore throat or a feverThe system uses machine learning algorithms to determine the right action for the userThe company says its new system can be used in a variety of home automation products, such as thermostats, door locks and door sensors.
The company is one of several in Australia using the technology, which has been developed by researchers at the University of Sydney and University of Queensland, to develop tools that are “smart enough” to help automate problems, according to the Australian Government.
“This is an area of very interesting research that I think is going to be of great value to Australian businesses,” said Paul Hickey, a senior lecturer in machine learning at the university.
“It will enable them to detect when a user needs help and take that step.”
Key pointsDiscovery, developed by University of NSW researcher, can detect when someone is sickThe technology is built on machine learning that is trained on real-world data from users and can be automatically used to predict when a problem may occurDiscovery has a number of different applications in the home automation market, including for door locks, thermostat sensors, and air conditioning systemsThe software is used in home automation devices such as door locks to help prevent people from leaving the house and for thermostatic devices to control the air conditioner.
“We’ve developed a system that works with our sensors to detect whether or not a user might have a condition like a sore, sore throat, sore fever, a cough, or even if they’ve got an allergic reaction,” said Professor Hickey.
“And if they do, we can tell the company that there’s a problem, which in turn can help the user take the necessary action.”
The company claims its discovery system can help solve some of the biggest challenges in home systems, including the “possible sudden onset of illness, a fever or a sore”, among other issues.
“In the future we may be able to use the discovery technology to assist in diagnosing people with serious medical conditions such as cancer or diabetes,” Professor Hipper said.
“Or we may use it to help find people with a problem with their personal data or personal privacy that’s important to them.”
The discovery technology was developed by a group of researchers at both universities, which include Dr Alex Linder and Dr Ian O’Keefe from the University’s Department of Electrical Engineering and Computer Science.
“There are many types of information systems that we can use for this type of task,” Professor Linder said.
The technology could be used for a number different applications, such like door locks that automatically detect when people leave the house, or thermostatically controlled air conditioners.
“The most important thing is that the system is capable of identifying the problem at hand and then providing a prompt response,” Dr O’Keefe said.
There is a number for each of the systems, such the system for door lock detection uses a set of real-life sensor data, and could be built for a range of home systems including a thermostaton system, a thermo-fence system, and a water heater.
“As well as the obvious applications, there are many others that we could build into our discovery system,” Professor O’ Keefe said, adding the company is working on adding more sensors to the technology.
“I think that the potential is pretty good.
It’s very scalable.
We’re trying to get a good number of sensors.””
It’s like a smart home that’s smarter than your house,” Professor John Liddell, a researcher in the same research group, said.
“You have sensors, you have sensors that you can control.
It can get a very intelligent response.”
The technology has been around for about five years, but the research was first reported by TechRadar.