Page 355 - Ai_V3.0_c11_flipbook
P. 355

• In supervised learning, the algorithm learns from labelled data, where each training example is paired with a
                      corresponding target label.
                       • There are two algorithms of supervised learning named regression and classification.
                       • Unsupervised machine learning algorithms are used when the information used to train is neither classified
                      nor labelled.
                       • Clustering and Dimensionality Reduction are common tasks in unsupervised learning.
                       • Regression  is  a  Machine  Learning  algorithm  used  to  analyse  the  relationship  among  dependent  variable
                      (target) and independent variable (predictor). It predicts the output values based on input values.
                       • Regression is basically used when the dependent variable is of a continuous data type. The independent
                      variables, on the other hand, can be of any data type—continuous, nominal/categorical etc.
                       • There are several types of regression analysis, random forest regression, support vector regression, decision
                      tree  regression,  linear  regression,  polynomial  regression,  ridge  regression,  lasso  regression  and  logistic
                      regression.
                       • Classification is a supervised learning concept which groups a set of data into classes.
                       • Classification is mainly of four types—binary classification, multiclass classification, multi-label classification,
                      and imbalanced classification.
                       • The KNN algorithm is a classifier using supervised learning and non-parametric learning (makes no assumptions
                      about the original data distribution) that employs closeness to classify or forecast the arrangement of a single
                      data point.
                       • Clustering, or cluster analysis, is the task of grouping a data set into similar items.
                       • Clustering is of four types, centroid-based clustering, density-based clustering, distribution- based clustering,
                      and hierarchical clustering.
                       • K-Means clustering is a powerful and widely used machine learning algorithm for partitioning datasets into
                      clusters based on feature similarity.
                       • The degree of association is measured by a correlation coefficient, represented by r. It is also called Pearson's
                      correlation coefficient and measures linear association between two variables.
                       • Causation shows that an event is the direct result of the occurrence of another event, i.e. a causal relationship
                      exists between the two events. This is also called cause and effect.






                                                              Exercise




                                                         Solved Questions

                                                   SECTION A (Objective Type Questions)
                          uiz
                 A.  Tick ( ) the correct option.

                    1.  Which of the following best describes the primary capability of machine learning?
                       a.  It allows computer systems to acquire knowledge and improve through experience without being
                         specifically programmed.
                       b.  It is a type of hardware that increases computer processing speed.
                       c.  It involves manually coding each function and rule that the system must follow.

                       d.  It is a method to directly control hardware using machine code.

                                                                                  Machine Learning Algorithms   353
   350   351   352   353   354   355   356   357   358   359   360