nate7378/Ethiopian-News-Information-Retrieval-System

# Ethiopian-News-Information-Retrieval-System The Ethiopian News Information Retrieval System aims to enhance access to Amharic news articles by implementing an efficient document retrieval framework. The primary objective was to develop a system that effectively indexes and retrieves relevant documents based on user queries, utilizing techniques such as text normalization, tokenization, stemming, and the BM25 ranking algorithm. The system was built using Python with Flask as the web framework, enabling a user-friendly interface for search functionality. A collection of Amharic news documents was processed to create an inverted index, which facilitates quick lookups and retrievals. Evaluation metrics, including precision, recall, F1-score, Mean Average Precision (MAP), Normalized Discounted Cumulative Gain (NDCG), and Mean Reciprocal Rank (MRR), were employed to assess performance.