Machine learning is the study and construction of algorithms that can learn from and make predictions on data. Algorithms follow strictly static program instructions by making data-driven predictions or decisions, through building a model from sample inputs. Machine learning is a subfield of computer science concerned with computational learning, evolved from the study of pattern and recognition, and computation learning theory in artificial intelligence.
Examples of Machine Learning
Microsoft Advanced Threat Analytics (ATA): Using its proprietary algorithm, Microsoft ATA works around the clock to pinpoint suspicious activities in your systems by profiling and knowing what to look for. ATA identifies known advanced persistent threats and security issues. ATA continuously learns from the behavior of users, devices, and resources and adjusts to reflect the changes in a rapidly evolving enterprise. As tactics get more sophisticated, ATA uses behavioral analytics to adapt and respond.
Tesla Motor Vehicles: Tesla currently uses a vision system provided by MobileEye. This technology is an advanced image-recognition system, capable of identifying road signs or obstacles such as other cars or pedestrians, on the road ahead. It uses deep learning, a popular machine learning technique based on training a many-layered network of simulated neurons to recognize input using many training examples. It also uses data from radar and ultrasound sensors and to make driving decisions.
Amazon: Amazon uses machine learning in several different scenarios. In the warehouse, it is used to package purchased items. Customer behavior and sales data is used to determine which products you’re most likely to also be interested in. Machine learning takes into account all of the existing data, from visits, clicks, and purchases, to predict behavior and determine recommendations that fit a customer’s particular interests. New products and new data are constantly being added to the system, so recommendation results are continuously adjusting and improving.
Are there any negatives to machine learning?
A machine is not human, but it does perform tasks that would be impossible for humans to accomplish manually. As with Tesla, machine learning algorithms are designed to perfection, but there’s still had a .01% chance of getting into an accident.
Why care about machine learning?
With the increase in IoT and connected devices, we now have access to countless amount of data and information. With it comes an increased need to manage and understand what we know about the data.
Further reading and resources