Go Back

Machine Learning

Machine Learning provides computers with the ability to learn from data without being explicitly programmed. Due to that, it can improve the state of knowledge workers’ decision making and/or completely automate big data discovery and execution processes. There are many factors driving the growth of machine learning:

1. High volume of data generated by sensors, controllers and machines
2. Complexities of connected subsystems and their interactions result in system dynamics that can no longer be fully comprehended, even by the smartest engineers
3. Realization that traditional system engineering has become a bottleneck in delivering costeffective solutions
4. Availability of less-expensive in-memory storage, faster compute hardware and easy-touse cloud solutions

Due to the above factors, digital industrial businesses will adopt machine learning in more and more use cases. For instance, in healthcare’s computer-aided diagnostics, machine learning models take as their input the patient’s condition —such as vital signs, symptoms, lab tests or toxic exposure—to provide disease classification or even recommended therapy. Adoption of computer-aided diagnostics has been low, driven by physicians’ resistance and the current lack of confidence society has in these models. This will change as evidence-based diagnostics continue to improve — to offer more accurate diagnosis than human judgments, given the enormous and ongoing proliferation of sensors as well as the use of Big Data to capture expert diagnoses and improve recommendations based on this knowledge base.