Page 148 - Ai_C10_Flipbook
P. 148

Evaluation Metrics for Classification


              Classification is a type of supervised learning in machine learning where the goal is to predict the categorical label
              or class of a given input based on historical data. In classification tasks, the model is trained on a labelled dataset,
              where a specific type of class label is the result to be predicted from the given input field of data. The model learns
              to map inputs to the correct category during the training phase.
              What is Classification?

              Classification is the task of “classifying things” into sub-categories. Classification is part of supervised machine
              learning in which we put labelled data for training. For example, You and your friends go to a restaurant, where
              pure vegetarians sit together at one table and non-vegetarians sit together at another table, to ensure that there
              is no confusion while serving food.
              So basically, you are classifying your friends into two categories:

              •  Pure vegetarians          • Non-vegetarians

                                     CLASSIFICATION IN MACHINE LEARNING










                                                                                        Vegetarian
                                                                                        Non-vegetarian
                                                                                        Eggetarian
                                                                                        Vegan
                                           4 Classes                2 Classes

              Classification usually refers to categorisation of a specific class label that needs to be predicted from the given
              input field of data as a result. For example, here we are working on a pure vegetarian service model that predicts
              whether the item served is for a pure vegetarian or for all.
              Classification Metrics


              When evaluating a classification model, it’s important to measure its performance using various metrics. These
              metrics help assess how well the model predicts the correct classes.
              Here are some popular metrics used for classification models:

              •  Confusion matrix     • Classification accuracy     • Precision     • Recall
                       Confusion Matrix


              A confusion matrix is a performance evaluation tool used in machine learning to assess the performance of
              a classification model. It is a tabular representation that compares the actual labels (true outcomes) with the
              predicted labels (model predictions). The table is made with 4 different combinations of predicted and actual
              values in the form of 2×2 matrix. The comparison between the prediction and the reality can be used to evaluate
              the rate of success. It also gives a clear picture of which classes are being predicted correctly and incorrectly, and
              the types of errors are being made.
              This matrix is also known as the Error Matrix and is used in situations where we need to evaluate the performance
              of the model, identify errors, and find ways to improve the efficiency of the model. It is useful for measuring Recall,
              Precision, Accuracy and F1 Score.

                    146     Artificial Intelligence Play (Ver 1.0)-X
   143   144   145   146   147   148   149   150   151   152   153