Page 30 - Ai V2.0 Flipbook C8
P. 30

Overview of Stages of AI Project Cycle


                  By defining the problem, gathering relevant data, analysing it, building a model, and evaluating
                  its performance, the AI project cycle helps ensure your project meets its objectives. This cycle
                  offers a structured roadmap for creating effective AI solutions.

                  The following figure shows the stages of AI Project Cycle:



                                                             Problem Scoping




                                             Deployment                       Data Acquisition










                                              Evaluation                      Data Exploration




                                                                Modelling




                  The description of the stages of AI project cycle is as follows:
                     • Problem Scoping: This is the first step where the problem is identified and clearly defined using
                     the 4Ws—Who, What, Where, and Why. It helps in understanding the issue and planning how AI
                     can solve it.
                     • Data Acquisition: In this stage, relevant raw data (text, images, videos, etc.) is collected from
                     sources like the Internet, books, or surveys to support the project.

                     • Data Exploration:  The collected data is analysed and visualised using statistical tools to
                     understand patterns, trends, and insights.
                     • Modelling: Suitable algorithms are selected to build and test models that can solve the problem
                     effectively. Different models are compared to choose the best fit.

                     • Evaluation: The model is tested to check its performance. If it fails to meet the goals, data or the
                     model may be adjusted. Once ready, the model proceeds to deployment.
                     • Deployment:  The  final  model  is  implemented  in  a  real-world  environment  with  ongoing
                     monitoring and maintenance to ensure continued success.
                  These stages help in systematically address challenges from the initial problem identification.

                  It ensures that data collected is relevant and complete, which contributes to building a reliable
                  model and gaining meaningful insights.




                           28     Touchpad Artificial Intelligence (Ver. 2.0)-VIII
   25   26   27   28   29   30   31   32   33   34   35