Page 186 - AI Ver 3.0 class 10_Flipbook
P. 186

Rule-based Approach

              The  Rule-based  Approach  is  one  of  the  earliest  and  simplest  methods  of  implementing  artificial  intelligence.
              It relies on predefined rules and facts created by developers to enable machines to perform specific tasks and
              generate desired outputs. Developers manually define a set of rules that determine how the machine processes
              data and responds to various scenarios.


                                    Rules
                                                      Rule-based Approach                 Answers
                                    Data

              The main drawback of this approach is that the machine's learning is static. Once trained, the machine does not
              adapt to changes made in the original training dataset. If the machine is tested on a dataset that differs from the
              rules and data provided during the training stage, it will fail to produce accurate results and will not learn or adjust
              to the new conditions it encounters.

              CASE STUDY: Banks' Chatbot

              A bank's website features a chatbot to assist users with basic queries related to account management, such as
              checking account balance, resetting passwords, or locating nearby ATMs.
              1.  Data: The data required to train the chatbot is a simple scripted document based on questions and their
                  corresponding answers.

              2.  Rules: The chatbot uses a simple decision tree approach with defined rules under the category of “Yes” or “No”
                  to complete the conversation. Some of the rules which bank's chatbots follow are:
                  a.   Rule 1: Predefined Questions and Answers. It prompts the user for “Account Balance”, if the answer is Yes, it
                    suggests a mobile app of the bank or login to an online banking account. Based on the user choice, it does
                    the needful.

                  b.   Rule 2: Keyword Matching. The chatbot recognises user input through keywords and matches them to
                    pre-set rules in its database. For example, when a user mentions "ATM," the bot asks for location details and
                    retrieves nearby ATMs from a stored database.
                  c.   Rule 3: Guided Flow Under the “Change Password” option. It will prompt the user to answer the details like
                    registered email or phone number. It will direct the user to the login page.
                  d.   Rule 4: Options and Redirection. Provides links to relevant pages (e.g., login, directions) and redirects to
                    human support for more complex issues.

              3.   Interaction: When a user communicates with the chatbot, it processes the message by matching it with the
                  predefined rules. Depending on the situation, the chatbot replies with a ready-made response or asks for more
                  details to address the query.
                  The following is an example of the user interaction:
                     ✶ User: "How can I check my account balance?"

                    Chatbot: "To check your account balance, you can use our mobile banking app or log in to your online
                    banking account. Would you like me to guide you to the login page?"
                    o  Option 1: Yes

                    o  Option 2: No




                    184     Touchpad Artificial Intelligence (Ver. 3.0)-X
   181   182   183   184   185   186   187   188   189   190   191