1MACHINE LEARNING Introduction Machine learning is used in various fields, the algorithms are very useful for various purposes, this essay describes regarding two vital algorithms that are used by the field of data science, and the algorithms are supervised ML algorithms and unsupervised ML algorithms (Meng, Bradley& Yavuz, 2016). This essay described regarding this in a detailed manner. It further presents a real world example of how artificial neural nets use supervised learning for predicting outcomes in decision making. Body Two main types of machine learning algorithms are supervised ML algorithms and Unsupervised ML algorithms. Supervised ML algorithms are the algorithms which involves the process of direct supervision of operation (Xiao, Rasul & Vollgraf, 2017). The developers are responsible for labelling sample data corpus along with setting boundaries that are very strict in nature and on these the algorithm operated. Most widely utilized supervised ML algorithms are logical regression, random forest, linear regression, neural networks and many more (Burrell, 2016). Unsupervised ML algorithms does not involve complete direct control of a particular developer. The difference among the two algorithms include the fact that supervised ML algorithms use labeled data in an exclusive manner whereas unsupervised ML algorithms learn feeds on various data that are unlabeled (Abadi, Barham & Chen, 2016). Most widely used unsupervised ML algorithms include K-means clustering, t-SNE, PCA and some more. Artificial neural nets make use of supervised learning for the purpose of predicting outcomes in decision making Neural nets are types of deep learning models that make use of large amount of training data in order to identify various correlations between various variables for learning to
2MACHINE LEARNING process incoming data in future. Artificial neural nets utilize the supervised learning for the purpose of predicting various outcomes in order to carry out better decisions, these outcomes are used by various organizations to predict the outcome of a particular strategy and hence they are able to undertake better decisions (Jordan & Mitchell, 2015). These helps an organizationincarryingoutaparticularprojectinaneffectivemanneralongwith maintaining a proper balance between the costs as well as resources incurred in the project. Assuming the outcomes helps the organization in not to proceed with a specific technique in case it does not provide benefit to the organization. One real world exampleof how each type of learning is applied in the field of data science Onereal world examplewhere supervised ML algorithms are used includes suppose a particular housing price predictor would not just take square footage, it would also take the overall number of bedrooms, overall number of bathrooms, number of floors, zip code, year built and many more similar factors (Obermeyer & Emanuel, 2016). Determining the inputs that would be used is considered as an important part of the supervised ML algorithms algorithm. The goal of supervised ML algorithms is not to carry out perfect guesses, this is because supervised ML algorithms deal in domains in which no similar factor exists (Meng, Bradley & Yavuz, 2016). The main goal is to make various guesses that are useful enough. Thereal life examplewhere supervised ML algorithms is used include credit card fraud detection. This is considered as one of the most useful advantage provided by the usage of supervised ML algorithms, based on various transactions made by numerous customers, the algorithm identifies which transaction is genuine and the one that is the result of a fraud transaction, this guesses are made based on numerous inputs that are provided by a customer (Xiao, Rasul & Vollgraf, 2017). In case a customer puts the pi n code in the first chance, it is
3MACHINE LEARNING considered as genuine and if the user puts the pin in multiple attempts, it is considered as fake transaction or fraud transaction. In case the use inputs a pin in the backside front or vice versa format it is also considered as a fraud transaction (Abadi, Barham & Chen, 2016). This way the multiple number of imputes provided by customers helps supervised ML algorithms in detecting weather a transaction is genuine or fake. Conclusion From the above discussion that had been carried out in this particular essay, it can be stated that the there are two main types of machine learning, these include supervised ML algorithms and unsupervised ML algorithms.This essay discusses regarding the artificial neural nets on how it uses superficial learning for predicting outcomes in the process of decision making, it also presents few real life examples of how machine learning is used in the field of data science.
Paraphrase This Document
Need a fresh take? Get an instant paraphrase of this document with our AI Paraphraser
4MACHINE LEARNING References Abadi, M., Barham, P., Chen, J., Chen, Z., Davis, A., Dean, J., ... & Kudlur, M. (2016). Tensorflow:Asystemforlarge-scalemachinelearning.In12th{USENIX} Symposium on Operating Systems Design and Implementation ({OSDI} 16)(pp. 265- 283). Burrell, J. (2016). How the machine ‘thinks’: Understanding opacity in machine learning algorithms.Big Data & Society,3(1), 2053951715622512. Jordan, M. I., & Mitchell, T. M. (2015). Machine learning: Trends, perspectives, and prospects.Science,349(6245), 255-260. Meng, X., Bradley, J., Yavuz, B., Sparks, E., Venkataraman, S., Liu, D., ... & Xin, D. (2016). Mllib:Machinelearninginapachespark.TheJournalofMachineLearning Research,17(1), 1235-1241. Obermeyer, Z., & Emanuel, E. J. (2016). Predicting the future—big data, machine learning, and clinical medicine.The New England journal of medicine,375(13), 1216. Xiao, H., Rasul, K., & Vollgraf, R. (2017). Fashion-mnist: a novel image dataset for benchmarking machine learning algorithms.arXiv preprint arXiv:1708.07747.