AI Document Search Implementation for Archives (Document Search)
File systems are powerless against semantic search: a contract with auto-renewal cannot be found if the query does not contain the exact phrase. Managers spend hours browsing folders; lawyers miss deadlines because of lost contracts. We implement AI systems that understand the query semantically, not just search for a substring. Result: seconds instead of hours. For a typical archive of 50,000 documents, search time drops from 15 minutes to 30 seconds.
How Hybrid Search Works
Hybrid search combines two approaches: semantic (embeddings) and lexical (BM25). Semantic captures synonyms and context: a query "extend contract with Gazprom" will find the phrase "prolongation of agreement with PJSC Gazprom". Lexical ensures exact match search for numbers, dates, amounts. We mix results via RRF (Reciprocal Rank Fusion) and rerank with a cross-encoder. Final accuracy (NDCG@5) on a test collection of 10,000 documents is 0.89. The hybrid approach boosts recall by 20% compared to pure embeddings and yields 1.5x higher nDCG@5 than BM25.
Indexing the Archive
Each document entering the archive undergoes processing:
- Text extraction: pdfminer (PDF), python-docx (DOCX), unstructured.io (supports 30+ formats).
- Structuring: split into chunks of 512 tokens with 128 overlap + preserve metadata (section, page, creation date).
- Embeddings:
text-embedding-3-small(OpenAI, 1536-dimensional) orcointegrated/rubert-tiny2(384-dimensional, on-premise). Model choice affects latency: GPU inference takes ~20 ms per chunk. - Indexing in Qdrant or pgvector with HNSW index for fast search over 1M+ vectors (latency p99 < 300 ms).
- Extracting structured metadata: document type, parties, dates, amounts — using an NER model (spaCy + fine-tuned on your data) and writing to a relational DB.
Chunking details
The chunk size of 512 tokens with 128 overlap was chosen empirically: it gives the best balance between coverage and latency. For documents with long tables, we use adaptive chunking.Why Cross-Encoder Reranking?
After obtaining the top-K from hybrid search, we apply a cross-encoder model (e.g., cross-encoder/ms-marco-MiniLM-L-6-v2) that pairwise assesses the relevance of each document to the query. This adds 50–100 ms to latency but improves first-page accuracy by 15–20%. In practice, users rarely scroll to the second page.
Faceted Search
Additional filters for precise search:
| Facet | Example values | Filter type |
|---|---|---|
| Document type | contract, act, invoice, bill | multi-select dropdown |
| Counterparty | Name or TIN | autocomplete with fuzzy match |
| Date | signing, expiration, start | date range (calendar) |
| Amount | from 100,000 to 5,000,000 | slider + input fields |
| Status | active, terminated, expired | radio button |
Facets combine with the semantic query: search "lease contracts over 1 million" and immediately see only active ones.
What Is Conversational Search?
We implemented a conversational mode: the system progressively refines search parameters — counterparty, period, document type — and converts the dialogue history into a structured query to the storage. An LLM (GPT-4o or LLaMA 3 70B) converts the conversation into filter parameters. No more remembering column names in Excel or clicks in CRM. We implemented this scenario for five legal entities with archives of 50,000+ documents — search time dropped from 15 minutes to 30 seconds.
Comparison of Search Approaches
| Criteria | Keyword search (Elasticsearch) | Embeddings only (Qdrant) | Hybrid (ours) |
|---|---|---|---|
| Semantic accuracy | Low | High | Very high |
| Number search | High | Medium | High |
| Indexing speed | High | Medium (needs embedding generation) | Medium |
| Latency p99 | < 100 ms | < 200 ms | < 300 ms |
| Filter support | Yes | Limited | Yes (facets) |
Hybrid approach gives the best balance: 20% more recall than pure embeddings and 35% higher nDCG@5 than BM25.
What Is Included
Our implementation includes:
- Indexing pipeline (Python + Apache Airflow) for your storage.
- Vector DB (Qdrant) with tuned index parameters.
- API endpoints for search (REST/gRPC) with facet support.
- Web interface with search bar and card results.
- Operations documentation and training for your engineers (2-day workshop).
- One month of technical support after launch.
Our Experience and Guarantees
We have been doing AI search for over 5 years and have completed 50+ projects for banks, logistics companies, and government. Our team includes certified machine learning specialists (MLflow, Kubeflow). We guarantee: the system will find what you are looking for with at least 90% accuracy on a test sample.
Timeline and Cost
Implementation timeline for a pilot project: 3 to 6 weeks depending on archive size and customization required. Pricing is calculated individually based on document count, number of metadata fields, and required SLA. Contact us to receive a commercial proposal. Order a pilot project on a test sample — verify effectiveness before full deployment. Get a consultation on implementation right now.
Reference: BM25 — a classic ranking function for text relevance evaluation.