Page 307 - AI Ver 1.0 Class 10
P. 307

6.  A company has built an AI model that gets 100% accuracy on training data. When they deployed this model on the
                       client-side it was found that the model is NOT at all accurate. What could have been the reason? There were no other
                       technical issues found at the client site.
                       a.  Wrong data was fed to the model               b.  Underfitting occurred
                       c.  Overfitting occurred                          d.  None of these

                    7. ................................ helps to analyse and interpret emotions in text data using text analysis techniques.
                       a.  NLP                                           b.  Summarization

                       c.  Sentiment Analysis                            d.  Text Classification

                 B.  Fill in the blanks.
                    1.  ………………………. are flexible, powerful, AI based models that have wider functionalities and support machine learning
                       algorithms that make a machine learn from the experience.
                    2.  The goal of ………………………. is to identify sentiments among several social media posts or even in a post where emotion
                       is not always explicitly expressed.
                    3.  ………………………. is the language used by humans to interact with the people around them.

                    4.  A ………………………. can be any word or number or special character that forms a part of a sentence.
                    5.  The process of removing the affixes from the words to get back its base word is called ………………………. .
                    6.  ………………………. is a simple and important technique used in Natural Language processing for extracting features from
                       the textual data.
                    7.   ………………………. is the process of dividing the sentences further into tokens.
                    8.  ………………………. technique is a statistical measure of words.

                    9.  ………………………. is the process of cleaning the textual data.
                   10.  ………………………. process of removing the affixes from the words to create a meaningful base word is  better than
                       stemming.

                 C.  State whether these statements are true or false.
                    1.  Snapchat is an example of NLP.                                                          ……………….

                    2.  Stemming is better than Lemmatization.                                                  ……………….
                    3.  BOW is considered a better technique than TFIDF.                                        ……………….

                    4.  Problem Scoping is a technique used in NLP for extracting information.                  ……………….
                    5.  BoW is a technique used for extracting features from the textual data.                  ……………….

                    6.  Creating a document vector is the collection of data for processing into Normalised corpus.   ……………….
                    7.  The frequency of a word in one document is called Document Frequency.                   ……………….

                    8.  A token can be any word or number or special character that forms a part of a sentence.   ……………….
                    9.  Stemming is a slower process.                                                           ……………….

                    10.  BoW algorithm creates a vocabulary of sentences.                                       ……………….










                                                                               Natural Language Processing  305
   302   303   304   305   306   307   308   309   310   311   312