Organizations in all industries are increasingly finding ways to leverage the information hidden in natural language to solve customer service issues, get insights from product reviews and medical records, and so much more. These insights depend on the ability of machines to understand and generate natural language. So how can your organization implement natural language processing (NLP) solutions for business problems in a smart, strategic way? With this report, department and data science team leaders walk through four enterprise use cases involving real companies. If you own the implementation of a project involving text analytics and NLP, you'll learn about the workflows, challenges, and key lessons learned that enabled each of these companies to get impactful results. Author Kinga Parrott, a senior AI strategy leader, discusses the critical importance of setting up the annotation process to transfer subject matter expertise to AI systems in NLP projects. You will learn when to set up the annotation process, how to effectively engage subject matter experts, and common pitfalls to avoid. Use cases in this report cover the following business needs: Once you review the use cases, you'll also learn steps for setting up your organization to tackle some of the most challenging and rewarding data science projects.
Table of Contents
- Teaching AI the Language of Your Business
- Introduction
- Use Case 1: Human Resources Document Management
- Business Need: Capturing, Managing, and Protecting HR Documentation
- Solving the Problem: Systems Architecture, Technologies, and Techniques Applied
- Challenges
- Key Success Factors and Lessons Learned
- Use Case 2: Enabling Virtual Agents to Answer Questions from Health Care Providers
- Business Need: Faster, Friendlier, and More Consistent Service to Providers
- Solving the Problem: Systems Architecture, Technologies, and Techniques Applied
- Challenges
- Key Success Factors and Lessons Learned
- Use Case 3: Reframing the Future of Due Diligence for Mergers and Acquisitions
- Business Need: Speeding Up the Transaction Diligence Process
- Solving the Problem: Systems Architecture, Technologies, and Techniques Applied
- Challenges
- Key Success Factors and Lessons Learned
- Use Case 4: Reducing the Legal and Financial Risks Associated with Service Contracts
- Business Need: Reduce the Legal and Financial Risks of Service Contracts
- Solving the Problem: Systems Architecture, Technologies, and Techniques Applied
- Challenges
- Key Success Factors and Lessons Learned
- How AI Understands the Language of Your Business
- Transferring Intelligence to AI
- Getting Annotation Right Reduces the Time to Value of NLP Projects
- The Annotation Process is Key and Must Be Collaborative
- NLP is Only as Valuable as the Results It Provides the Business
- Introducing Errors throughout the Pipeline
- Iteration
- Acknowledgments