Name
Tutorial/Best Practices: Amplifying Human Content Expertise with Real-World Machine-Learning Workflows
Date & Time
Tuesday, October 25, 2022, 3:00 PM - 3:30 PM
Plamen Minev
Description
Human-led content classification and enrichment has long been the most impactful, yet most expensive form of content workflow operations. Large content library owners often find that they have irreplaceable content expertise in only a few contributors that becomes a critical gating factor in how quickly a large content library can be transformed from a pile of ingest project and tape files to become rich content that is fully annotated, searchable, and enhanced for modern user consumption.
With AI machine learning tools, it is possible to leverage that human expertise, then unleash the power of AI for content processing workflows that can work tirelessly on even the largest content libraries and performed within the framework of commonly used workflow tools such as an asset management system for the highest efficiency of operations.
In this session, we will take the user through the approach, steps and processes of taking a time-consuming, novel content application task normally performed by a human expert, adapt that task to a machine learning challenge, build and deploy the necessary ML model, then apply it to content as a part of ongoing content operations using familiar asset management systems.
The session will also offer recommendations to avoid many of the pitfalls of machine-learning exploration undertaken by many brands and content producers and explain how to leverage the team’s shared expertise to keep models and libraries up to date.
Objective:
Session attendees will gain:
• An understanding of the steps and processes of selecting, and defining a human-led task for suitable replication by machine learning libraries
• A referenceable workflow example integrated with an asset management system and resources within the content owner’s facility that leverages cloud-delivered tools, or a hybrid solution
• Best practice recommendations on unifying content-owner machine learning activity, AI/ML tools for ongoing organizational efficiency
Location Name
Salon 1
Take-Aways from this Presentation
Readers will gain: • An understanding of the steps and processes of selecting, and defining a human-led task for suitable replication by machine learning libraries • A referenceable workflow example integrated with an asset management system and resources within the content owner’s facility that leverages cloud-delivered tools, or a hybrid solution • Best practice recommendations on unifying content-owner machine learning activity, AI/ML tools for ongoing organizational efficiency