DocLens.ai edge - platform differentiators
Risk scorecard
Proprietary risk assessment scores with the flexibility to incorporate specific user input at Configuration stage
Guided by expertise
Data labeling of training sets done by expert lawyers and claims experts; all models thoroughly manually reviewed for accuracy
Know your risk AI Engine
AI solution built from the business user's perspective; uses state of the art NLP techniques; high accuracy models to drive business outcomes
Insights & 3rd party data
Repository drives the application of insights and risk mitigation strategies consistently across the entire portfolio; drives benchmarks and standards. Specific 3rd party data augments risk engine and drives benchmarking
Data and AI
At DocLens.ai, we use Artificial Intelligence (AI) to help professionals including lawyers save time while reviewing hundreds and thousands of documents. Our solution hinges on:
1
Natural Language Processing: (NLP) is a subfield of artificial intelligence (AI) concerned with processing and analysis of natural language data to develop an "understanding" including the contextual nuances in the content. This technology can be used to accurately extract information and insights contained in legal documents such as contracts or insurance claims
  • Legal Question Answering: we ask specially trained NLP models legally pertinent questions, for example what is the governing law, to identify relevant answers in legal documents such as contracts. The foundational AI model used and modified is licensed under a CC BY 4.0 license. The modifications made by DocLens.ai are not covered by that license
  • Named Entity Recognition(NER): we identify specific legally relevant entities such as the Venue in Claims files using pretrained models on NER tasks
  • Legal Document AI: applies different NLP AI methods and text search techniques with legal phrases. We use small to large sized state of the art AI models ranging from a few million (BERT) to more than a billion parameters. We use open source Python frameworks such as Huggingface Transformer models to rapidly prototype new approaches. We employ latest hardware accelerated AI model retraining and inferencing to custom build solutions to meet our enterprise clients unique needs
2
Legal Risk Understanding: our team includes legal experts who evaluate the results of our AI solution and design innovative ways of inferencing presence of unfavorable provisions in legal documents
3
Secure Private Cloud Infrastructure : legal documents almost always contain sensitive personal information and potential trade secrets. We set up governed, secure and compliance ready dedicated cloud infrastructure for our enterprise customers