Page 371 - AI Ver 3.0 Class 11
P. 371

3.  Learning: Learn and understand the meaning of facts related to the topic. Gather and arrange facts supporting your
                    argument and formulate your arrangement each time when new evidence arrives. This will help you find updated or
                    completely new information that can score points against your opponent’s position.
                 4.  Interacting:  Listening  to  your  opponent’s  arguments  and  opinions,  then  present  a  convincing  altercation  that
                    further proves your case.







                   Understanding              Reasoning                 Learning                   Interacting
                   Cognitive systems          They reason underlying    They never stop learning.   They interact with
                   understand like humans do.  ideas and concepts. They   They develop "expertise"   humans.
                                              debate, infer, and extract   with every interaction and
                                              concepts.                 outcome.

                                 Task                                                       #Creativity



                     Find five interesting facts about “IBM Project Debater” and write in the space provided below.   [CBSE Handbook]

















                        Natural Language Processing (NLP)

                 Computers excel in working with structured data, in which everything is properly organised and labelled. Unfortunately
                 for machines, human language is not structured. You've spent most of your life communicating through human language.
                 Your brain accomplishes this with some of the most complex neural circuitry on this planet. However, creating machines
                 that can understand human language, is extremely challenging.
                 NLP  involves  machines  segmenting  sentences  and  extracting  meaning  from  “tokens”  of  human  language.  Human
                 language  is  difficult  and  unstructured.  Despite  being  loosely  held  together  by  grammar  constraints;  our  language
                 presents information in a variety of puzzling ways. Unstructured information, unlike structured information that can
                 be  organised in tables or matrices with clearly labelled rows and columns, is disorganised and difficult to comprehend.
                 Consider the statements made by famous personalities:
                 •  Groucho Marx: “One morning I shot an elephant in my pajamas. How he got into my pajamas I'll never know.”
                 •  Steven Wright: “I poured spot remover on my dog. Now he’s gone.”
                 •  George Carlin: “I went to a bookstore and asked the saleswoman, ‘Where’s the self-help section?’ She said if she
                    told me, it would defeat the purpose.”
                 To deal with the “messiness” of unstructured data, computers start with one sentence at a time. The process is known
                 as sentence segmentation that is sentence-based text analysis. In NLP analysis, the text data is either analysed using
                 meaningful words called tokens, or analysed using sentences. Computers divide the input into small piece of information
                 (tokens), which may be classified separately. NLP can deal with text tokens after they have been sorted into a structured
                 format or manner based on their meaning.
                                                                   Leveraging Linguistics and Computer Science  369
   366   367   368   369   370   371   372   373   374   375   376