Build a classical information retrieval system first, then extend it with grounded answer generation.
Evidence
Evaluation
The repository README reports these Cranfield ranking scores. They are shown as repository-reported values, not recomputed for this portfolio.
Condition: Cranfield document-ranking evaluation
These scores describe retrieval ranking only; they do not measure generated-answer quality or grounding. NDCG@1,400 covers the full ranked set recorded in the existing project notes.
Problem
RAG systems are easier to market than to evaluate. This project starts with a classical IR system on Cranfield, then adds a generation layer so retrieval quality stays measurable instead of disappearing behind an LLM response.
Role / contribution
Project author responsible for dataset parsing, indexing, retrieval, evaluation, and the RAG extension described in the repository.
Architecture flow
What happens
Cranfield documents, queries, and relevance labels are extracted into a structured IR pipeline.
Why it matters
A structured benchmark pipeline is what makes retrieval evaluation possible from the start.
Related tools
Engineering decisions
Decision 01
The project does not hide retrieval behind the generation layer; evaluation remains a first-class output.
Decision 02
The RAG extension is layered on top of a measurable baseline rather than mixed into the core retrieval logic.
Decision 03
Cranfield gives the project a concrete evaluation frame instead of anecdotal search results.
Technical implementation
The core engine parses Cranfield, preprocesses text, builds an inverted index, and constructs a sparse TF-IDF retrieval pipeline ranked with cosine similarity.
Evaluation is explicit through MAP and NDCG, which keeps retrieval behavior inspectable before generation is layered on top.
The RAG extension passes top retrieved documents into Gemini so answers stay connected to the retrieval output rather than free-form generation.
Stack
Limits and next steps
Project explainer
A case-study guide to the architecture and decisions behind Mini Search Engine with RAG Extension.