Sifting through content with AI at KMWorld 2019

When so much information is coming into the organization it’s a challenge to sift through it. AI, autoclassification, and text/document analytics can improve how businesses analyze and govern knowledge.

At KMWorld 2019 Katrina B Pugh, faculty, Information and Knowledge Strategy Master of Science Program, Columbia University and AlignConsulting, and Sandra Serkes, president and CEO, Valora Technologies and Columbia University IKNS, related hard-earned lessons learned about scoping, cost-justifying, and adopting these technologies during their session, “AI: Virtual Agents & Content.”

According to Pugh and Serkes, content bloat costs “1 hour, 41 minutes/week spent searching (average worker) and 13.2 billion hours across all US workers.”

It is imperative that companies take control of unstructured data to adhere to compliances, identify how important the data is, and disposition, Pugh and Serkes said.

Videos of KMWorld 2019 awards presentations,  keynotes, and sessions can be found here.

Metadata is the frame, it tells us what it is, what it’s about, gives off inherited labels, impact and risk. Auto-classification can analyze and design that data. The solution contains software algorithms that determine the disposition of content.

Once data is identified for what it is, it needs to go through the information governance lifecycle, Pugh and Serkes explained.

The data need to go through processing, tagging, disposition, and business assessment.

“There’s good metadata and suspect metadata,” Pugh said. “Knowing what documents you have, its meaning, redundancy, and what risks you have means being able to protect it.”

KMWorld 2019 is a part of a unique program of five co-located conferences, which also includes Enterprise Search & Discovery, Office 365 Symposium, Taxonomy Boot Camp, and Text Analytics Forum.

Many speakers at KMWorld 2019 are making their presentations available at www.kmworld.com/Conference/2019/Presentations.aspx.


KMWorld Covers
for qualified subscribers
Subscribe Now Current Issue Past Issues