It is used to draw inferences from datasets consisting of input data without labeled responses. Unsupervised learning methods find hidden patterns or intrinsic structures in data. We could also predict the weather condition whether it is going to snow or not tomorrow by a binary classification problem. In this situation, we could predict tomorrow’s temperature by a regression model. For instance, historical temperature data, amount of precipitation, wind, snow, and humidity. In order to make correct predictions for the weather, we need to consider various inputs. Now, we could train a supervised learning model to predict a new house’s price based on the examples observed by the model. We then need to know the prices of these houses by leveraging data coming from thousands of houses, with their features and prices. If there is data about the houses, such as the square footage, number of rooms, features, whether a house has a garage or not, and so on.

Supervised learning uses classification and regression techniques to develop predictive models. A supervised learning algorithm takes a known set of input data and known responses to the data (output) and trains a model to generate reasonable predictions for the response. The goal of supervised learning methods is to build a model that makes predictions based on evidence in the data. Machine learning uses two types of techniques: supervised learning, which construct a model on known input and output data so that it can predict future outputs, and unsupervised learning, which finds hidden patterns in input data. Supervised Learning Methods vs Unsupervised Learning Methods Sample size calculation or power analysis is directly related to the statistical technique that is chosen, because the sample size calculation is based on the power (typically 0.80 is desired), and the effect size (typically a medium or large effect are selected the larger the effect, the smaller a sample is needed). Non-parametric techniques must be used for categorical and ordinal data, but for interval & ratio data they are generally less powerful and less flexible and should only be used where the standard parametric test is not appropriate―e.g., when the sample size is small. Besides, the researcher should have good knowledge of parametric methods and non-parametric methods. Moreover, the type of data is also a fundamental concept in the analysis, for example the techniques appropriate to interval and ratio variables are not suitable for categorical or ordinal variables. In addition, the researcher should have clear idea of the variables that will be used in the research work, whether they are categorical or nominal, ordinal, or rank-ordered, interval, or ratio-level. In order to overcome such problems, the researcher should be aware of the major differences between possible statistical modeling approaches that could be applied simultaneously. In some research situations, there could be some confusion on choosing the most appropriate technique for the analysis, because different techniques seem to be applicable. The wrong selection would lead to incorrect interpretation of the results and inadequate findings.

Selection of the correct statistical approach is vital in any research work.

Decision tree vs random forest