lawTechCamp - Knowledge Management Panel

Leveraging Knowledge to Increase Responsiveness

Ron Carriere, CEORon Carriere, CEO 2011 2011

Cognitive-Intelligence-AppPresented at

LawTech Camp 2011

Cognitive Informatics' Research Institute

Cirilab Inc.

2000 – 2011 Cirilab Incorporated in 2000.R&D on Advanced Multi-Dimensional Indexing Technology for Text Clustering and Profiling. Privately Owned Research Center to Develop Advanced Knowledge Management (KM) Technology Products.

The Problem

• “information overload” was identified as the business problem of 2008 and estimated to be a $650 billion drain on the US economy. Companies like Microsoft, Google, Cisco, and Oracle are all working to try to solve the problem that they’ve helped create. Something’s got to give.

• According to a study by research firm Basex and reported by The New York Times,

The Source •   a folder of documents

• · the contents of a web site

• · all the posts in a Blog

• · all the posts in a Discussion Group

• · unstructured text records within a Customer Relations Management system

• ·  and many other unstructured text collections.

• The “Captions” from a video file

The ROI

• “Did you know the average information worker spends up to 10 hours a week searching and gathering information from the enterprise? Do you know how much that costs your organization? According to IDC, over $15,000 per employee with an average salary of $75,000.

META-Knowledge  

• Soft computing differs from conventional (hard) computing in that, unlike hard computing, it is tolerant of imprecision, uncertainty, partial truth, and approximation. In effect, the role model for soft computing is the human mind. The guiding principle of soft computing is: Exploit the tolerance for imprecision, uncertainty, partial truth, and approximation to achieve tractability, robustness and low solution cost.

The Solution

Let The Documents Express Themselves

• And give me a way to access them based on what I am interested in, not what some search engine wants me to see! Cirilab takes a very different approach. As with Cirilab Speed Read, we allow each document to express its own thematic content through a Knowledge Signature™. With a Cirilab Knowledge Map™ we simply combine all of the Knowledge Signatures™ of individual documents into one thematic hierarchical representation of the entire collection. A Cirilab Knowledge Map™ can be created

from almost any repository of unstructured text

Review Information Faster

• Cirilab’s software not only finds content across multiple documents but also helps you review content when you find it.

0100020003000400050006000700080009000

Number of Words

K-Sig Themes Synopsis DetailedSummary

EntireDocument

Document Type

Information Triage

Words

Autonomous Knowledge ObjectsAutonomous Knowledge Objects

•Self-Organizing

•Self-Simplifying

•Self-Describing

KOS

KOS

KOS

Technology ComponentsKGE

Speed Read

Enterprise Server K-store

Knowledge Map

Yahoo Discovery

Research Workbench*Web Ready PublishingKnowledge

Generation EngineTM

Your InterfaceHere via API!

List of functions

• Read a document to identify and summarize relevant information

• Profile and organize a document collection or website

• Index and build a knowledge base of all your documents

• Discover on the index to build a research knowledge base

• Cleanup by finding duplicates and similar documents

Fast decision making using Cirlab’s Fast decision making using Cirlab’s Intelligence systemsIntelligence systems

Databases

MinutesPress Releases

PoliciesOfficial Records

Workshop/Meetings

Correspondence

Active Organizational Intelligence (index)Active Organizational Intelligence (index)Index

Knowledge Query By:• Document• Expert profile• Natural language

Technical manuals

Regulations

Legal Documents

Policies

Project Reports

Correspondence

Official Records

Internet

Intranet

Query Organizational IntelligenceQuery Organizational Intelligence

Databases

Discover

Semantic Wave

19

Knowledge Value Chain

s

Strategic

Operational “

Tactical

MCSI BI- Text Analytics- Semantic Intelligence -Business Intelligence-Meta- Cognition

- Knowledge Monitoring- “What do they know”- Functional Reporting

- Text Monitoring and “What are they talking about”

Information Vectors Information Systems

What are they talking about?

Situation

• The Leukemia & Lymphoma Society is the world's largest voluntary health organization dedicated to funding blood cancer research, education, and patient services.

• The web site’s Discussion Boards have 32 categories and over 108,000 Posts.

• There is a desire to understand conversation patterns within common discussion themes to better understand how web site users are communicating

Approach• LLS will extract all 108,000 Discussion Board posts

– Each Discussion Board category will be stored in one Folder– Each Topic Thread within a Discussion Board will be saved as

one HTML file that will contain the initial post as well as all of the following post replies.

– The Topic Thread title will be used as the file name within the Category folder

• Cirilab’s Thematic Extraction and Knowledge Mapping software to create a Thematic Knowledge Map Profiles of each of the 32 Discussion Board categories

• MauroNewMedia will use the Knowledge Maps to analyze the thematic similarities within and among Topic posts and across DB categories

Summary

• Cirilab’s Thematic Extraction technology looks for common “thematic veins” across multiple documents within a collection and allows the researcher to quickly locate them.

• Cirilab’s Discover technology allows the analyst to search available information for additional insight to “thematic veins” that have been discovered.

• Contact info: Ron Carriere

• Rcarriere@cirilab.com or Mhurst@cirilab.com

• WWW.Cirilab.com

• Note Webinars available