Page 86 - DCAP506_ARTIFICIAL_INTELLIGENCE
P. 86

Artificial Intelligence




                    Notes             Static representation — knowledge regarding objects, events, etc., and their relationships
                                       and states specified.

                                      Needs a program to recognize what to do with knowledge and how to perform it.
                                   Procedural Representation

                                   In Procedural representation, knowledge is signified as procedures.


                                          Example: Default reasoning and probabilistic reasoning  are examples of  procedural
                                   techniques.
                                   Here, heuristic knowledge of “How to do things efficiently “can be simply signified.
                                      Control information essential to use the knowledge is entrenched in the knowledge itself.
                                       For example, how to discover relevant facts, make inferences etc.
                                      Needs an interpreter to follow instructions provided in knowledge.


                                          Example: Let us assume what knowledge an alphabetical sorter would want:
                                      Implicit knowledge that A occurs before B etc.
                                      This is simple really integer comparison of (ASCII) codes for A, B.

                                           All programs enclose procedural knowledge of this sort.
                                      The procedural information here is that knowledge of how to alphabetise is signified explicitly
                                       in the alphabetization procedure.

                                           A declarative system might have to have explicit details such as A occurs before B, B
                                            comes before C etc.

                                   6.1.1 Representing How to Use Knowledge

                                   We discuss below the need to represent how to control the processing:
                                   Direction

                                      Specify the direction an allegation could be used. For example, to verify something can fly
                                       show it is a bird. fly(x)   bird(x).
                                   Knowledge to Achieve Goal

                                      Identify what knowledge might be required to attain a specific goal.


                                          Example: To prove  something is  a bird  attempting using  two facts  has_wings  and
                                   has_feathers to demonstrate it.
                                   Actually most of the knowledge representations utilize an amalgamation of both. Most of the
                                   knowledge representation structures have been generated to manage programs that  control
                                   natural language input. There are numerous types of schemas that have proved functional in AI
                                   programs. They comprise:
                                   (i)  Frames: Used to explain a compilation of attributes that a specified object possesses (e.g.:
                                       depiction of a chair).
                                   (ii)  Scripts: Used to illustrate general sequence of events (e.g.:- a restaurant scene).




          80                                LOVELY PROFESSIONAL UNIVERSITY
   81   82   83   84   85   86   87   88   89   90   91