LegalTech Law-as-a-Service (LaaS)
Acknowledgements
- Agile/Scrum
- Open-Source
- Cloud Service
- 3rd Party Tools
- Webinars
- Portals
- Repositories
- EdTech
Global Legal Market SWOT Analysis
(The Business Research)
Entities: offices of lawyers, title abstract and settlement offices and offices of notaries
Legal professionals: solicitors, barristers and lawyers
Value $767.1 billion in 2021
CAGR of 5.4% since 2015
CAGR of 6.5% to $986.7 billion by 2023
Asian Industry Demand (40% of companies in the Fortune 500):
automobile, technology, oil and gas and e-commerce
Emergence Of Alternative Legal Service Providers (ALSPs) linked to LA Market
high-demand legal services such as document review, contract management, litigation support, electronic discovery, contract lawyers and staffing, legal research, and IP management
(from $10.7 billion in 2017 to $55 billion by 2025),
Big Four accounting firms – Deloitte, EY, KPMG, and PwC
Legal Analytics (LA) Driven Digitalization (LexisNexis Surveys)
70% cost reduction in legal information reviews
72.4% of the legal firms are supporting cloud adoption
B2B legal service opportunities
$117.2 billion of global annual sales by 2023, and $48.0 billion in the USA
Regulatory Challenges in OECD Countries
LegalTech Market Share SWOT Analysis
Value $740.14 Million in 2019
$6,625.95 Million by 2027
CAGR of 31.5% 2020-2027
Analytics Type: Prescriptive, Predictive, Descriptive
Case Type: IP, Commercial Case, Anti-Trust Management
Geography: North America, Europe, Asia Pacific
Key Players:
IBM, Wolters Kluwer, Mindcrest,
Thomson Reuters, Unitedlex,
Lexisnexis, Argopoint
End-Users:
Law Firms, Corporates,
Government, SMEs and Others
cf. BusinessWire Report:
$456.1 Million in 2017
$5100.3 Million by 2026
CAGR of 30.8%
What is LA/LaaS Ecosystem?
Why LA/LaaS?
Case Assessment & Prediction1: Assessing Litigation Cases and Case Outcomes
Judge Analysis1: Highlight Relevant Arguments of Judges in Similar Cases
Opposition Scrutiny1: Build Profiles of Opposition using Open-Source Case Logs
Client & Witness Insights1: Build Profiles of Clients/Witnesses using Public Sources
Case Strategy Development1: Quick Access to A Set of Facts Relevant to the Case
Win the Case2: Reveal Patterns in Data about Prior Litigation and Opposing Counsel
Attracting New Clients2: Advantage with Marketing/Business Development
Pricing Projects2: Price Legal Services more Aggressively and Remain Competitive
Cost Savings2: E-Billing, Fast QA Guideline Compliance, Reduced Administration Time
Transforming Effectiveness/Agility in Legal & Compliance (Fujitsu CoE)
Ex1: Improving Law Interpretability with NLP (StrangeLoop, 2019)
Ex2: NLP Optimized Search of Legal Documents (PyData, 2018)
Ex3: GCP Dialogflow Chatbot/Agent Semantic Analysis of Text Docs
Sources: 1Quanovo, 2LexisNexis
Deloitte Example: 16 AI Projects
A Virtual Legal Research Assistant TAX-I: Processed 1153 Tax Cases of the ECJ
An AI Benchmark Study of Transfer Pricing: Clusters of “Similar” Companies/Prices
SONAR: Find Labelling Errors in Databases: QC Processed 30000 Retail Products
WKR+VAT Analytics (ML+NLP Transaction Detector): Over 40 Clients
GRAPA Risk Strategy Assistance, Optimized Audit (High-Risk Anno Accounts PoC 2018)
Chatbot Search Tool Online Technical Library, Q&A User-Library Dialog Flow (cf. Ex3)
Argus Cognitive Audit Tool Scans and Reviews Contracts (e.g. Consulting/Tax)
24/7 Virtual Assistant Q&A Decision Trees (Chatbots, Smart Speakers, etc.)
Insurance Policy Risk Advisor Maps Out Policyholders with a Higher Risk of Losses
Predicting Payment Behaviour – Determine the Highly Reliable Value of the Debt
DocQMiner Self-Learning Contract Analysis (Brexit Re-Papering, US GAAP Compliance)
Eagle Eye (Web EWS PoC 2018): Internet Monitoring for Different Markets/Countries
Financial Advise: Combating Welfare Fraud, Unacceptable SAP ERP Transactions
BrainSpace e-Discovery Clustering Gathers the Evidence from Millions Text Docs
State-of-the-Art
Legal Tech Portfolio
Legal Analytics (LA) Concepts
Descriptive = reports/dashboards
Diagnostic = data discovery/mining
Predictive = ML-based predictions
Prescriptive = scenario testing
LA UI Products
Lexis, Westlaw, Logi, Wolters Kluwer, etc.
Statistical BI Analytics Software
SAS/SPSS, Tableau, MS Power BI, etc.
Open-Source BI Software
Rx64 3.6.2, Excel VB, Tableau Public, QlikView, etc.
Cloud Analytics
TIBCO Spotfire, Data Studio, IBM Cognos, Zoho, etc.
Cloud ML/DL/AI (Python)
Scikit-learn, TensorFlow, pyTorch, mxnet, Gluon, etc.
Cloud NLP/CX
GCP DialogFlow, Gluon NLP, Azure Cognitive API, etc.
Legal Tech Agile/Scrum/HR Analysis*
LA Scrum Team: Scrum Master, Product Owner, Data/BI/Scientist/Analyst/Engineer/Architect
Network: Legal Staff, Marketing (CMO), CRM, UX Research, C&P, HR, CFO, HSSE, DPAs
Programming/Scripting: Python (54%), R (45%), SQL (36%), MATLAB (19%), Java (18%), C++ (8%)
OS: Windows/Linux, DevSecOps Tools, RDBMS/NoSQL, Tableau/Power BI, Statistics (Excel, SAS)
LA Team Goals Explained:
Help Data-Driven Decision Making,
Build Analytics Models to Help Business Strategy,
Extract Key Business Performance Indicators
BI Analyst: Analyse Past Historical Data (e.g. Revenues), Dashboards/Graphs, KPIs, Competition
Data Analyst: Gather I/O Data, Structuring DB, Creating/Running Models, Explain Patterns
Data Engineer: Design/Support BI Platforms
Data Cleaning, Building/Maintaining ETL Pipelines,
Write/Analyse/Debug SQL Quieries,
Collecting/Managing/Analysing/Visualizing Large Datasets (e.g. Hadoop)
Data Architect: DB Architecture, DWH, Data Governance, Data Visualization (Tableau)
Create DB from Scratch,
Dataflow Management,
Data Storage Strategy
SLA-Oriented Integration
* Indeed.com, Glassdoor.com, Adzuna.com,
Jooble.com, EURES, legal500.com, etc.
Key LaaS Components
Descriptive LA – Trends, Averages, STD, Box Plots, X-Plots, Hist, etc.
Diagnostic LA – Hypothesis, Conjectures, Assumptions
Predictive LA – Regression Analysis, Logistical Regression
Prescriptive LA – Decision Making (Concept of Probability)
Unsupervised ML – Untrained Data Mining
Classification, Regression, Clustering
Supervised ML (CNN, NLP, Word Embedding, etc.)
Labelled Training Data + Test Data
Training – Test Predictions – Model Evaluation – Predictions
Deployment into BI (Azure Qlik or Tableau)
Reinforcement Learning (RL) – Software Agent Actions
Keras, Tensorforce, RL_Coach, ChainerRL, MushroomRL
Deep Learning (DL)
ML: parse data, learn from that data, and make informed decisions based on what it has learned
DL: create NN layers that can learn and make intelligent decisions on its own (e.g. via backpropagation loop)
On/Off-Policy
Quality of Actions
LaaS On-Prem/Virtual Workspace
Windows/Linux OS, cygwin, PuTTY SSH
Python 3 Libraries:
NumPy, matplotlib, scikit-learn, Kmeans, pandas, TensorFlow, Keras, PyTorch, NLTK, MXNet , etc.
R CLI + Packages for Data Science (mlr3, XGBoost, Caret, ggplot2, data.table, dplyr, tidyr, etc.)
Full-Stack Portfolio: Back-End (Java, C++, etc.), Front-End (JS, HTML, TS, etc.)
IDE (Jupyter, Eclipse, R/Android Studio, MS Studio), CLI (Linux Bash/sed, PowerShell)
SQL/PostgreSQL, NoSQL, Oracle RDBMS, MySQL DB Service, Graph DB (Neo4j), MS Access
User-Customized Open-Source/Commercial LA Products and Data Visualization UI
Agile/Scrum Project Management (Monday.com, PM, Jira, Hive, etc.), DevOps CI/CD Tools
Miscellaneous (e.g. Google Analytics)
Cloud Computing (CC) LA Virtualization
On-Premises, Public/Private/Hybrid/Multi-Cloud Environment
IaaS, PaaS, FaaS/Serveless, and SaaS
Highly Variable Cost (e.g. AWS 5 times more expensive than Azure for Windows SQL Server)
Significant Savings through Existing Licenses (up to 49% for Azure)
Main Services: AI, Analytics, VM, DB, Developer Tools, DevOps, IM, IoT, BI Reports, etc.
IaaS: IT Infrastructure Outsourcing Model (e.g. CPU/GPU)
PaaS: Tools for Developing Applications (e.g. Kubernetes Container Orchestration)
FaaS: Serveless application development environment as microservices at lower cost than PaaS
SaaS: 3rd Party Licensed Applications - Gmail, Office 365, SalesForce (CRM), PayCom (HR), QuickBooks (Finance)
MPI HPC (e.g. Broadcast, Linux Clusters, Intel)
Big-Data Processing (Spark/Hadoop)
VMs, Kubernetes and Docker Containers (PaaS)
AWS, IBM, Azure, and GCP offer VPCs via VPN
Edge Locations (e.g. 31 Azure Edge Sites)
Industry 4.0 Legal Challenges - Best CC Practices
Disadvantages: Dependence, Security, Data Mobility
Default: Bare-Metal Computing (No Virtualization)
Comments
Post a Comment