Page 145 - Artificial Intellegence_v2.0_Class_12
P. 145

Stage  6: Data Preparation:   his  stage  contains  all  the  activities  to  build  the  dataset  used  in  the  subsequent
                     modeling stage. Activities to prepare data include
                     •   data cleansing  handling missing or invalid values, removing duplicates, applying correct formats),
                     •    oining data from multiple sources  files, tables, platforms), and
                     •   the conversion of data to more useful variables.
                     Data preparation is usually the most time consuming procedure in a data science pro ect. Automating certain data
                     preparation steps in advance can speed up the process by minimising ad hoc preparation time.
                 .    hat are hyperparameters   hat is their purpose   ive e amples of fe  hyperparameters.
              Ans.     yperparameters are parameters  hose values govern the learning process.  hey are 'top level' parameters that
                     regulate the learning process and model parameters that come from it, as the prefi  'hyper' suggests.  ince the
                     model cannot modify its values during learning training, hyperparameters are said to be e ternal to the model.
                      ome e amples of hyperparameters are
                     •  he ratio of train test split

                     • Optimisation algorithms' learning rate  e.g. gradient descent)
                     •  he loss function that the model  ill employ
                     • A neural net ork's number of hidden layers
                     • A clustering task's number of clusters
                 .     plain the purpose of evaluation and deployment stage.
              Ans.     valuation  tage
                      he data scientist
                     •  utilises a testing set for predictive models  that is separate from the training set but follo s the same
                        probability distribution and has a kno n outcome.)  he testing set is used to assess the model and ad ust it as
                        necessary.
                     •  or a final assessment, the final model is sometimes applied to a validation set as  ell.
                     In addition, data scientists can use statistical significance tests to verify the model's accuracy.

                     Deployment  tage
                      he model is deployed into the production environment or an equivalent test environment once it has been built
                     and authorised by the business sponsors. It is usually used in a restricted capacity until its effectiveness has been
                     thoroughly assessed. Deploying a model into a live business process frequently necessitates the involvement of
                     additional internal teams, skills, and technology.

                             ased/Application-b
                                               ased questions:
            C.  Competency-based/Application-based questions:
                       ency-b
               Compet
                 .    onsider the follo ing statements containing an assertion and a reason
                    elect the appropriate option for the statements given above
                   a.    oth A and   are true and   is the correct e planation of A
                   b.    oth A and   are true and   is not the correct e planation of A
                   c.   A is true but   is false
                   d.   A is  alse but   is true

                    i.   Assertion (A): At the core of every AI model is ‘finding patterns in data.’
                       Reason (R):  inding the right pattern is usually an iterative process.
                 Ans.  b
                   ii.   Assertion (A):  onsider that the goal of an AI model is to predict an ans er such as "yes" or "no".
                       Reason (R): In such a case, predictive modeling can be used.
                 Ans.  c

                                                                                       C apstone  P roj e ct
   140   141   142   143   144   145   146   147   148   149   150