Discover game-changing breakthroughs in knowledge management
A renewed enthusiasm about knowledge management is evident, particularly in the digital workplace. Discovering the value in corporate knowledge is amplified by new technologies, such as text analytics, cognitive search, machine learning, natural language processing, mobile email management, and new policy efforts.
To explore game-changing breakthroughs that are expanding what is possible with knowledge management, KMWorld recently held a webinar featuring Val Swisher, founder, Content Rules; Chip Gettinger, VP, global solutions consulting, SDL; and Al Stevens, director, clinical content management, EBSCO Clinical Decisions.
EBSCO Industries (The Elton B. Stevens COmpany) is a diverse organization with many businesses, including information services, real estate development, insurance, hunting and fishing equipment, and more, Stevens explained. EBSCO Information Services is the leading provider of research databases, e-journals, magazine subscriptions, e-books, and discovery service to libraries of all kinds, and EBSCO Clinical Decisions is a leading healthcare industry provider of clinical decision support solutions, healthcare business intelligence, medical journals, medical e-books, and medical reference information.
The company has been able to add and bookmark information regarding COVID-19 by using structured content and DITA. It is taxonomy driven and editor published information. A knowledge graph is key to sorting this information.
The organization’s knowledge graph contains hundreds of thousands of nodes and several million terms, Stevens said.
By utilizing machine learning the company can deliver content with context.
“We use machine learning using standard supervised learning models,” Stevens said. “We train on 100 topics categorized by editorial as: newborn, infants, children, adolescents, adults, elderly. ML classification accuracies of 80%+ enables us to effectively use the patient age to prioritize results.”
Structured content is the backbone of creating a cognitive system, Swisher said. Cognitive Systems do not innately “know” anything, they must be trained on data.
Cognitive systems determine the best answer to a question based on statistical significance. If you have more than one instance of the same content, the statistical significance can be diluted. Uploading structured content that contains a single source of truth is one of the best ways to create and then curate your corpus, Swisher explained.
The content is curated by human experts. During curation, experts discard information that is out of date, immaterial, and deemed to be no good. After curation, the content is ingested. When an AI Natural Language Processor ingests content, it pre-processes the content, creating indices and metadata. This makes working with the content more efficient in the future.
Once ingested, human experts train the AI on how to interpret the information. This is called machine learning.
Gettinger took a different approach to addressing these technologies. He recommended SDL Tridion which can intelligently create, manage, and deliver business-critical information—at a global scale.
Businesses risk lost revenue or worse if customers can’t find the right information, Gettinger said. SDL Tridion can dynamically drill down to the right information. Its search predicts the ideas a user is looking for.
Intelligent delivery needs taxonomy tagging, said Gettinger. Upcoming editorial features enable authors to easily tag and architects govern with instant tagging suggestions users can work with industry taxonomies or build their own and then approve, govern, and localize taxonomies.
An archived on-demand replay of this webinar is available here.