Page 308 - Artificial Intellegence_v2.0_Class_11
P. 308

A ‘false negative’ is dangerous for the patient—the test says you don’t have the disease while you actually suffer from
              that disease, especially in diseases like cancer or HIV Aids, Covid-19.



                             Brainy Fact

                    “Confusion Matrix’ is so called because we want to see whether the system is confusing between the two
                    classes—Reality and Prediction. Confusion matrix is even used in Psychology for the identification of stimuli!



                                                                                                   Experiential Learning

                          Video Session


                     •   Scan the QR code or visit the following link to watch the video: Machine Learning
                        Fundamentals: The Confusion Matrix
                        https://www.youtube.com/watch?v=Kdsp6soqA7o
                     •  Scan the QR code or visit the following link to watch the video: Making sense of the confusion matrix
                        https://www.youtube.com/watch?v=8Oog7TXHvFY
                       After watching the videos, find out what is true positive rate (TPR), true negative rate (TNR), false-positive
                     rate (FPR), and false-negative rate (FNR).






                      Clustering
              Suppose you plan to go on a holiday to Amritsar, Punjab. You plan to visit around 20 tourist spots in Amritsar in three
              days. How will you do this? Write down your plan.

              You will most probably group the tourist spots into say 3 – 5 groups. The ones closer to each other like the Golden
              Temple and Jallianwala Bagh will go in one group, while Wagah Border which is further away will be included in another
              group. You do this so that you can park your car in one spot/central location and walk around visiting all the tourist spots
              in your group/cluster.
              The above can be classified as a clustering problem. Clustering, or cluster analysis, is the task of grouping a data set into
              a set of similar items. It is an unsupervised learning algorithm. As an important analysis method in machine learning,
              clustering is used to identify patterns and structures in unlabelled data sets. Clustering algorithms are used in many
              fields, as given below.


              Applications of Clustering
              There are various applications of clustering. Let us learn about some of them.

              Fake News Detection
              The algorithm uses the content of fake news articles, called a corpus, checks the words used, and then groups them into
              clusters. These clusters help the algorithm determine which is real and which is fake news. If you see a high percentage
              of certain terms in the article, the material is more likely to be fake news.








                    306     Touchpad Artificial Intelligence (Ver. 2.0)-XI
   303   304   305   306   307   308   309   310   311   312   313