It is derived from the concept that it deals with the construction and the study of system that can learn from data. According to Arthur samuel , 1959 the machine learning is the subfield of computer science that gives the computer the ability to to learn without being explicitly the programmed
A computer program is set to learn from ETP. Where E stands for experience T stand for task performed with experience P for performance , measure the performance of the task and improves with it E Experience . Machine learning can be seen as building blocks to make computer to learn to perform more efficiently or intelligently
Although machine learning is subfield of computer science but it may vary from traditional approaches of computer science . In traditional ways algorithms are sets of overtly programmed commands used by computers for problem solving or calculating .
Machine learning and Technology:
Machine learning algorithms in its place let computers to train on data inputs and use statistical analysis in order to output values that fall within a specific range. just for this, machine learning facilitates computers in building models from sample data in order to automate decision-making processes based on data inputs. This machine learning has benefited the technology user . Facial acknowledgement technology permits social media platforms to help users tag and share photos of friends. Optical character recognition technology changes images of text into portable type. Recommendation engines, power-driven by machine learning may propose what movies or television shows to watch next based on user partialities. Car navigations are also available for users .
Categories of machine learning:
In machine learning, chores are generally categorized into broad four categories. These categories are based on how learning is acknowledged and how the feedback is given on the learning of the developed system. These are the the four major categories or methods of learning.
In supervised learning the correct classes of trained data are known it depends on algorithm trained by human input condensed expenditure on manual review for application and coding. Therefore supervised learning therefore uses patterns to calculate label values on additional unlabeled data.
In unsupervised learning the correct classes of trained data are not known or the data is unlabeled and in contrast there is high reliance on raw data with large expenditure .it is used for transactional data it provides the algorithm with no labeled data in order to allow it to find structure within its input data. Let’s explore these methods in more detail.
Semi supervised learning
Semi supervised learning is the combination of both supervised and unsupervised learning. it stressed on analytical trained by human input and calculated analysis.
Reinforcement learning allows the machines or software to improve or learn from the feedback the change in behavior from feedback may be for once or keep up going with respect to feedback.