Welcome to the Pet Projects repository. This repository contains personal projects that I have initiated and implemented on my own. These projects reflect my personal interests and experiments in various domains.
Description: The Azerbaijani Legal Acts Search Engine (ALASE) leverages state-of-the-art NLP techniques to provide users with an intuitive interface to search and retrieve pertinent legal acts from the extensive Azerbaijani legislation corpus. Utilizing a distilled version of BERT, known as DistilBERT, for encoding both the corpus and user queries, ALASE offers unparalleled accuracy and speed in returning search results that are semantically aligned with user intent.
Technologies Used:
- DistilBERT
- Natural Language Processing (NLP)
- Python
- Django
- Elasticsearch
- Other relevant libraries (e.g., Transformers, Hugging Face, etc.)
Highlights:
- Implemented DistilBERT for efficient encoding of the corpus and user queries
- Achieved high accuracy and speed in returning semantically relevant search results
- Developed an intuitive user interface for easy interaction with the search engine
- Indexed an extensive corpus of Azerbaijani legislation
Link to Notebook/Code: NLP Semantic Search Code