Supervised and Unsupervised Prediction Application of Machine Learning

Machine learning field is introduced at the level of concept. Ideas such as supervised and unsupervised as well as regression are explained. The trade-off between bias, diversity, and the complexity of the model is discussed as the primary study guide concept. The different types of models that can be produced by machine learning are introduced as a neural network (feed and repetitive), vector support machine, random forest, map maps, and Bayesian network. Model training is discussed next with its main ideas for dividing the database into sets of training, testing, and validation and performing cross-validation. In this chapter, we begin by reviewing the basics of machine learning such as feature testing, unregulated and supervised learning and classification types. Then, we identify the main problems in designing a machine learning test and evaluating its performance. Finally, we introduce supervised learning methods. Modeling the quality of the model is handled next to the important role of the site specialist in keeping the project real. The paper concludes with some practical tips on how to create a machine learning project.