Machine Learning and Big Data

7Machine Learning

Elham Ghanbari* and Sara Najafzadeh

Department of Computer, Islamic Azad University, Yadegar-e-Imam Khomenini (RAH) Shahre Rey, Tehran, Iran

Abstract

Inspired by algorithmic and computational learning theory, machine learning examines the study and construction of algorithms that can learn and predict based on data—such algorithms do not just follow program instructions and Through modeling of sample input data, they predict or decide. Machine learning is used in computational tasks where the design and programming of explicit algorithms with good performance are difficult or impossible; some applications include email filtering, Internet intrusion detection, or internal malware intended to cause information breaches, learning ratings, and machine vision.

Keywords: Machine learning, concept learning, version space, supervised learning and unsupervised learning

7.1 History and Purpose of Machine Learning

7.1.1 History of Machine Learning

Machine learning exceeds the scope of artificial intelligence. In the early days of developing artificial intelligence as a science, the researchers found that machines learn from the data. They tried to solve this with a variety of symbolic methods and what then called “neural networks”; these methods were often perceptron and other models that were later determined to redesign the generalized linear models [1].

However, the growing emphasis on logical and knowledge-based methods made a gap between artificial intelligence ...