LEX — AI Legal Platform for Law Firms

AI-powered legal analysis platform for law firms and corporate counsel.

Features

Resources

Blog Articles

Technology

Built on AWS (EC2, Bedrock Claude AI, ALB, WAF, S3, ACM, KMS). PostgreSQL, Redis, Qdrant vector database. TypeScript, React, Node.js.

Start free — 50 credits on registration. Sign up

LEGAL 5 min

Searching Court Decisions by Meaning, Not by Keywords

You search for "compensation for apartment flooding" and miss the case where the court writes about "tortious liability for property damage resulting from engineering infrastructure failure." Keywords find words. Semantic search finds meaning.

Searching Court Decisions by Meaning, Not by Keywords

Keywords find words. Semantic search finds meaning.


Why the Court Decisions Registry Is Not Enough

The Unified State Register of Court Decisions (EDRSR) is an invaluable resource. But its search works on keywords. This means:

Example: You search for "apartment flooding." Case 753/12847/21, where the court writes about "tortious liability for property damage resulting from engineering infrastructure failure" — will not be found. Not a single word in common.

How Semantic Search Works

Instead of comparing characters, the system compares meaning:

  1. Your query is converted into a mathematical vector (embedding)
  2. Every decision in the database already has its own vector
  3. The system finds decisions that are similar in meaning, even when the words are completely different

Practical Examples

| Your Query | Keyword Search | Semantic Search | |———–|—————|—————–| | "apartment flooding" | Decisions with the word "flooding" | + "tortious liability for property damage" | | "eviction from mortgaged apartment" | Decisions with "eviction" + "mortgage" | + "foreclosure on pledged property" | | "rent debt" | Decisions with "rent" + "debt" | + "recovery of rental payments", "tenant arrears" |

What This Means for Practice

Research completeness. You find relevant practice you would never have found with keywords. Not 30 decisions — but 200-300, including those where the court used different terminology.

Speed. Instead of 10-15 keyword combinations — one natural-language query. The system finds all phrasing variations itself.

Non-obvious connections. Semantic search can find decisions from adjacent practice areas where the court applied an analogous legal approach. You would never have searched for it — but it is exactly what you need.

Keyword search answers the question "where are these words?" Semantic search answers "where was this kind of problem resolved?"