Legal Precedent Finder - USA

How is the legal data indexed and embeddings created?

• The data for legal precedents is collected from various U.S. legal databases, encompassing a wide range of case laws.
• The collected data is then normalized and standardized to ensure uniformity and accuracy.
• The text from the case laws is then extracted, preprocessed to remove any special characters or irrelevant elements, and tokenized.
• The cleaned text is then indexed using a custom indexing algorithm that considers the hierarchical structure and nuances of legal texts.
• OpenAI’s text-embedding-ada-002 model is used to create embeddings, transforming the text into high-dimensional vectors with a dimension of 1536.
• These vector embeddings are then stored in a database optimized for storage and retrieval of high-dimensional data.

How do we keep the data updated?

We maintain active connections with various U.S. legal databases to ensure that we have access to the most recent legal precedents. Our system continuously monitors changes in case laws, and any new rulings are collected, processed, and indexed. This ensures our embeddings are up-to-date, accurately reflecting the current legal landscape.

How to use

• Python
• Curl

Python Code

You can access the legal precedent embeddings using the embedelite Python package or by making a CURL request.

# This returns a list of embeddings
from embedelite import load_embedding

embeddings = load_embedding("legal-precedents-us")
print(len(embeddings))
print(embeddings[1])

# This returns an object that can be directly inserted into the database
result = load_embedding("legal-precedents-us", embed_for="embedelite")
# The result is {"embeddings": [], "documents": [], "ids": []}
print(result["embeddings"])
print(result["documents"])
print(result["ids"])

Curl

Request

curl -X POST -H "Content-Type: application/json" -d '{
  "doc_id": "legal-precedents-us"
}' https://api.embedelite.com/v1/embeddings/download/

Response

{
  "mappings": {
    "properties": {
      "doc_source": { "type": "keyword" },
      "sentence": { "type": "text" },
      "embeddings": { "type": "dense_vector", "dims": 1536, "index": false },
      "doc_source": { "type": "keyword" }
    }
  }
}