LexNLP Features

Information Extraction

Legal Terms

Extract Legal Terms

Built to find legal domain-specific text:

  • Find dates like effective dates, termination dates, or delivery dates
  • Find parties like persons and organizations
  • Find durations like terms, notice periods, or assignment delays
  • Find references to citations or regulations like “26 U.S.C. 501,” “26 CFR 31.3121,” or “655 F.3d 1013”
  • Find references to courts or regulators
  • Find controlling law and jurisdictions
  • Find copyrights and trademarks
  • Find definitions like “Default shall mean…”
Information Extraction

Financial Terms


All financial terms, regardless of regional or jurisdictional variants:

  • Find monetary and non-monetary amounts
  • Find and resolve currency symbols and abbreviations
  • Find and resolve geopolitical entities and distances 
  • Find percentages and rates
  • Find ratios and proportions
Natural Language Processing

Tokens and Segments

Tokenizes and Segments Text

Other libraries struggle with headings and sections, or they stumble over common abbreviations like “U.S.C.” LexNLP tokenizes and segments, leaving nothing behind:

  • Pages
  • Paragraphs
  • Sentences
  • Tokens, stems, and stopwords
  • Titles
  • Articles and sections
  • Exhibits and schedules
  • Table of contents
Natural Language Processing

Cluster and Classify


LexNLP can cluster and classify data in intuitive ways:

  • Cluster sentences, paragraphs, clauses, sections, and documents
  • Classify documents as legal or non-legal material
  • Classify documents as financial or non-financial material
  • Classify contracts among various contract types
  • Classify clauses among various clause types
  • Use deep semantic models to find synonyms or related concepts

Input Text

LexNLP – Extracted Values

Below is a sample of results from LexNLP using a restricted English legal term set.