These are the sources and citations used to research Machine Learning. This bibliography was generated on Cite This For Me on
Artificial Intelligence is considering the concept of machines having the capability of carrying out tasks that we have humans would consider as intelligent. Whereas Machine Learning is a branch of Artificial intelligence which is the idea that we as humans should be able to give machines access to data and information so that they can learn the tasks for themselves.
In-text: (Forbes.com, 2018)
Your Bibliography: Forbes.com. (2018). What Is The Difference Between Artificial Intelligence And Machine Learning?. [online] Available at: https://www.forbes.com/sites/bernardmarr/2016/12/06/what-is-the-difference-between-artificial-intelligence-and-machine-learning/#6ecb206b2742 [Accessed 7 Feb. 2018].
To create a good machine learning system, it will need: • Data preparation capabilities • Automation and iterative processes • Scalability • Ensemble modelling (Running two or more related but different analytical models and then synthesizing the results to a single score) • Algorithms
In-text: (Sas.com, 2018)
Your Bibliography: Sas.com. (2018). Machine Learning: What it is and why it matters. [online] Available at: https://www.sas.com/en_us/insights/analytics/machine-learning.html [Accessed 7 Feb. 2018].
Machine Learning can be broken down into different learning methods (algorithms) such as: • Supervised Machine Learning Algorithms – The machine takes data that it has learned from previous sessions and applies it to new data to predict future events. From analysis of a dataset, the algorithm produces an inferred function to make guess about the results. The learning algorithm can then compare its results with the correct results, so it can find any errors to rectify them. • Unsupervised Machine Learning Algorithms – These are used when the information used to train the machine isn’t classified or labelled. It researches how systems can infer a function to detail the hidden structure from unlabelled data, meaning the system doesn’t figure out the correct output, it researches all the data, so it can draw inferences from datasets to describe the hidden structures from unlabelled data. • Semi-supervised Machine Learning Algorithms – These algorithms are between supervised and unsupervised because they both use labelled and unlabelled data for training. Semi-supervised is used when the data requires skilled resources so that it can train the machine and learn from it. • Reinforcement Machine Learning Algorithms – This method interacts with its surroundings by producing actions and discovers errors or sometimes rewards. Trial and error is most common in reinforcement learning which allows the machines and software agents to determine the ideal behaviour with a context to have its performance at the highest standard. Delayed reward is another characteristic that reinforcement learning has. Simple reward feedback is used by the agent to identify what action is best, better known as the reinforcement signal.
In-text: (Expertsystem.com, 2018)
Your Bibliography: Expertsystem.com. (2018). What is Machine Learning? A definition - Expert System. [online] Available at: http://www.expertsystem.com/machine-learning-definition/ [Accessed 7 Feb. 2018].
10,587 students joined last month!