aws presentation memory

Justification of cloud-based

• The cloud scales automatically with workloads.

• It speeds up development and deployment.

• It has integrated services to make real-time data easy to manage.

• Less infrastructure meaning lower costs and less to manage.

Security, Identity, and Compliance

• IAM Roles to assign temporary permissions.

• Amazon Cognito manages user sign-in and authentication.

• AWS Shield protects our system from DDoS attacks and keeps our services running safely.

Management and Governance

• Amazon CloudWatch monitors services, tracks usage, checks for errors, and collects logs to help us manage everything.

additional categories

• We use Amazon SageMaker with an XGBoost model to predict fish sightings and activity, using real-time and historical data.

• Amazon Rekognition helps analyze images and videos to detect fishing events.

• Amazon Kinesis Video Streams collects live video feeds for real-time monitoring.

why we chose XGboost

XGBoost fits our fish AWS project because we have a lot of real-time and historical data that’s organized into fields like fish sightings and activity. XGBoost is really good at finding patterns in this kind of structured data, handles large datasets fast, and makes accurate predictions about when fish might be at risk. It's better than Linear Learner because it can catch more complicated trends, which is important when ocean conditions change a lot.

Operational Excellence

we monitor with CloudWatch, use three-tier architecture for reliability, and protect data with S3 backups.

FISHFACE

FishFace uses AI to identify and count fish after they are caught, helping with data collection. While it improves catch reporting, it only reacts after fishing happens. Our solution is better because we monitor fish populations live with IoT camera and use machine learning to predictions  before overfishing occurs. This protect ecosystems early, not after the damage is done.