Page 119 - Ai_V1.0_Class9
P. 119

Evaluate model performance using different metrics to ensure accurate detection and classification of plastic waste.
                  Validate models with cross-validation techniques and real-world testing in different ocean regions.

                    Deployment: Integrate the validated AI models  into the operational  environment for real-world
                    applications.  Deploy  models  using  scalable  and  efficient  frameworks,  monitor  performance
                    post-deployment, and iterate as necessary to maintain effectiveness and relevance.


                  Integrate AI models  into autonomous drones or underwater robots for real-time monitoring and cleanup
                  operations.  Implement  cloud-based solutions  for scalable deployment across  global ocean  regions.
                  Continuously monitor model performance and refine algorithms based on feedback from cleanup operations
                  and environmental assessments.
                  By following this AI project cycle, we can effectively leverage technology to address the critical issue of ocean
                  plastic pollution, leading to improved marine ecosystem.
                  Let us map the steps of Cleaning Ocean project to the steps in the AI project cycle.

                       Define                        Analyse data                    Evaluate model
                     objectives for                   to visualise                    accuracy and
                    AI-based ocean                   plastic waste                    effectiveness
                   plastic detection                 patterns and                        using
                     and removal.                    distributions.                  environmental
                                        Data                             Data           metrics.
                                     Acquisition                      Modelling                       Deployment
                      Problem                           Data                             Data
                      Scoping                        Exploration      Develop AI      Evaluation
                                    Gather satellite,                 models for                      Implement AI
                                      drone, and                       accurate                     models in drones
                                    oceanographic                     plastic waste                   for real-time
                                    data on plastic                  detection and                   ocean cleanup
                                     distribution.                   classification.                   operations.


                  Watch the video-The Ocean Cleanup
                  https://www.youtube.com/watch?v=xz21cKgRHxI






                         Problem Scoping and Setting Goals for an AI Project


                 Problem scoping is the term used to define the process of selecting a problem that we might want to solve using
                 AI knowledge. Identifying a problem and then having a vision to solve it is called problem scoping. Let us start
                 scoping a problem. Look around, we are surrounded by problems, big or small. At times we don’t feel the problem,
                 as we are so used to them. Look around and select a theme that interests you the most from the following diagram:



                                  Environment         Travel       Entertainment     Education
                                                     Tourism





                                             Cyber           Women             Social          Security
                                            Security          Safety          Welfare




                                                                         AI Reflection, Project Cycle and Ethics  117
   114   115   116   117   118   119   120   121   122   123   124