Discover how one of the world's largest law firms is revolutionizing its operations through advanced analytics in this 40-minute conference talk. Explore Clifford Chance LLP's journey from zero to data science, handling millions of legal documents annually for global clients. Learn about the challenges faced in the legal industry and how the firm's Data Science Lab is addressing them through innovative solutions. Gain insights into the architecture, client relationship analysis, and document classification techniques employed. Delve into the technical aspects of data cleaning, model development, and hyperparameter optimization using tools like Spark, NLP, and Azure Databricks. Understand the impact of these advancements on the legal industry and how they are reshaping business operations.
Revolutionizing Legal Industry with Spark, NLP and Azure Databricks - Clifford Chance Case Study
Databricks via YouTube
Overview
Syllabus
Intro
About Mirko Bernardoni
About Michael Seddon
Clifford Chance
Digital is changing how business gets done
What we do in the Data Science Lab
Al Adoption MATURITY CURVE
DS-Lab end to end capability
Confidentiality
Many challenges in legal
Architecture
Client relationship analysis
Document classification
Document example
Data download and cleaning
Models development · Cross validation with 10 datasets of 1000
Grid search on single core model
How does miflow manage multithreading in Spark?
Bayesian optimization for hyperparameter optimisation
Hyperopt example
Hyperparameters results
What results?
Taught by
Databricks
