Data Driven Model for Anomaly Detection and Path Prediction July 2022 - April 2023
Investigated and enhanced the Automatic Identification System of cargo vessels for anomaly detection and path prediction.
Technical stack Used in the Project -
- Explored a dataset related to AIS marine activity, with data recorded between January 1, 2022, and January 15, 2022, featuring a total of 17 unique attributes.
- Formulated a statistical method to analyze cargo ship vessels by considering time difference and speed over the ground for robust anomaly detection.
- Engineered a path prediction algorithm using a sequence-to-sequence model with an attention mechanism, delivering a substantial 90% reduction in error compared to Deep LSTM and GRU models.