Page 118 - Ai_V1.0_Class9
P. 118
Task #Creativity & Innovativeness
Case Study: Cleaning Oceans
The world's oceans are increasingly polluted with plastic waste, which poses a severe threat to marine life
and ecosystems. Identifying the location and density of plastic waste patches is challenging due to the vast
and dynamic nature of the ocean. Traditional methods of detecting and removing plastic waste are inefficient
and labour-intensive. Coastal communities and environmental organisations lack the advanced tools needed to
monitor and clean up these plastics effectively, leading to continued environmental degradation and harm to
marine biodiversity.
Can we solve this problem with AI? How?
Now that you are aware of AI concepts, plan to use them in accomplishing your task. Start with listing down all
the factors which you need to consider saving the oceans. This system aims to:
Problem Scoping: Define the problem statement, objectives, and constraints of the AI project. Identify
stakeholders, understand the business context, and outline the scope of work to be undertaken.
Define the scope to develop AI models for identifying and removing ocean plastic waste. Identify key challenges
such as ocean dynamics, plastic detection limitations, and scalability of cleanup efforts.
Data Acquisition: Gather relevant datasets required for training and evaluating AI models. Ensure data
is collected from reliable sources, adheres to legal and ethical standards, and is prepared in a format
suitable for analysis.
Gather satellite imagery, drone footage, and oceanographic data from various sources to create a comprehensive
dataset. Utilise data from environmental organisations, governments, and research institutions.
Data Exploration: Explore and analyse the acquired data to understand its structure, quality, and
relationships. This phase involves data cleaning, visualisation, and statistical analysis to uncover patterns
and insights that inform subsequent modelling.
Clean and preprocess the data to extract relevant features such as plastic concentrations, ocean currents, and
marine life habitats. Explore visualisation techniques to understand spatial and temporal patterns of plastic
waste.
Modelling: Develop and implement AI models using appropriate algorithms and techniques. This
phase includes feature selection, model training, parameter tuning, and validation to optimise model
performance and accuracy.
Develop AI models such as convolutional neural networks (CNNs) for image recognition to detect and classify
plastic debris in ocean images. Implement reinforcement learning algorithms for optimising cleanup strategies
based on real-time data.
Evaluation: Assess the performance of trained AI models using evaluation metrics and validation
techniques. Validate model predictions against ground truth data to ensure reliability and generalisation
ability across different scenarios.
116 Artificial Intelligence Play (Ver 1.0)-IX

