Project repo for building retrieval engine based on famous Algorithms like BM-25, Query-Likelihood etc
-
Updated
Mar 4, 2019 - Python
Project repo for building retrieval engine based on famous Algorithms like BM-25, Query-Likelihood etc
Subjectivity-aware RAG pipeline for investment questions with web search, multi-strategy scraping, Worker + Checker LLMs, and deterministic claim verification.
Information Retrieval with Lucene and CISI dataset. Index documents and search between them with IB, DFR, BM-25, TF-IDF, Boolean, Axiomatic, LM-Dirichlet similarity and calculate Recall, Precision, MAP (Mean Average Precision) and F-Measure
An AI-powered public grievance and complaint management platform built on the MERN stack. Features automated LLM-based priority routing, semantic search, and real-time dashboards for citizens and administrators.
Production-hardened agentic RAG: multi-agent supervisor, Corrective RAG, hybrid search (BM25 + RRF + Cross-Encoder + multi-granularity), OTel tracing, per-tenant rate limiting, A/B testing. 10 LangGraph patterns.
Information Retrieval Project : Crisis Facts Summarization
A simple full-text search engine on C++ with ranging BM25, factor TF-IDF. ITMO SE'27 first cource programming laboratory work.
Design, Implementation and Evaluation of various Search Engines
RAG over UK planning policy, built from scratch — hybrid retrieval with measured evaluation.
A privacy-first, zero-cost Hybrid RAG architecture featuring hierarchical auto-merging retrieval (Dense + BM25), in-memory PII scrubbing, and Groq LLM generation.
High-precision record and entity resolution framework leveraging hybrid lexical-semantic search and deterministic LLM verification to minimize false positives.
Applying SBERT and LightGBM for Detecting Duplicate Questions in Quora Website (Binary Text Classification)
To associate your repository with the bm-25 topic, visit your repo's landing page and select "manage topics."