Page 402 - AI Ver 3.0 class 10_Flipbook
P. 402

CASE STUDY: Sentiment Analysis For Internet Movie Database (IMDb)

              IMDb—The Internet Movie Database (IMDb) is an online database containing information and statistics about
              movies. IMDb was first published in 1990 by Col Needham, a computer programmer, as a group of scripts which
              allowed users to search a list of film credits that were compiled with the help of a USENET group. The database
              was incorporated in 1996. In 1998, Internet Movie Database Ltd. became a subsidiary of Amazon, which uses it as
              a way to advertise movies and DVDs which it sells on its main website.

              The sample data taken for our steps can be download from link.
              https://orangewebsupport.co.in/AI/IMDB_Dataset_ODM.xlsx
              OR
              http://bit.ly/4jLNdhG
              Steps to teach AI model to predict the sentiments at different stages of Project life cycle using the Orange Data
              Mining tool are as follows:

               Stage 1   Problem Scoping

              From the IMDb dataset, predict for each review in the corpus.
              Predict the following sentiments from the dataset:

               Sentiment          Meaning
               Positive Score     Positive Review

               Negative Score     Negative Review
               Neutral Score      Neutral Review

               Compound Score     Total Sentiment of the review i.e. -1 means most negative, +1 means most positive

               Stage 2   Data Acquisition

              Follow the following steps in the process of data acquisition:

               Step 1   Open the Orange Data Mining application and click on the New option.
                        It opens the new canvas.
































                    400     Touchpad Artificial Intelligence (Ver. 3.0)-X
   397   398   399   400   401   402   403   404   405   406   407