A custom Named Entity Recognition (NER) model built to analyze customer comments. It uses an LSTM-based architecture to detect key entities relevant to e-commerce engagement.
The model detects and tags the following entities:
- PRODUCT (e.g., sneakers, hoodie, watch)
- COLOR (e.g., red, navy, silver)
- LOCATION (e.g., Cairo, Dubai, London)
- DATE (e.g., tomorrow, friday, today)
Perfect for automating the analysis of comment sections to understand what users want to buy and where they are located.