MIS – Lecture-19

N P Singh
Learning Objectives
 What is the role of knowledge
management systems in business?
 What are artificial intelligence (AI) and
machine learning? How do businesses use
AI?
 What types of systems are used for
enterprise-wide knowledge management,
and how do they provide value for
businesses?
 What are the major types of knowledge
work systems, and how do they provide
value for firms?
What is Knowledge
Management
 “Knowledge management is
leveraging relevant intellectual
assets to enhance organizational
performance.”
AI Beats Radiologists in
Reading Mammograms (1/4)
 According to American cancer society
268,600 new cases of invasive breast
cancer (BC) and 41,760 deaths from BC
were predicted during 2019.
 As per the Globocan data 2020, in India, BC
accounted for 13.5% (178361) of all cancer
cases and 10.6% (90408) of all deaths.
 World data : 2 million cases new cases and
½ million deaths per year
 Leading Tool for detecting BC is
mammography Based on X-rays to locate
and diagnose tumors
 AI Beats Radiologists in
Reading Mammograms (2/4)
 US: 33 million screening mammograms are performed/year.
 Test has 20% false negatives & 7-12% false positives.
 New system of reading mammograms using AI is developed.
 Proving better than radiologists in reading mammograms
 A team of Google health, Northwestern University Chicago,
& British Cancer Research Imperial center & Surrey county
Hospital designed the computers model to read X-ray
images.
 US-False negatives – 9.4% & false positives 5.7%
 In UK: US-False negatives –2.7% & false positives 1.2%
 Training of model was done with data of 76000 women in UK
& 15,000 in US
 Test data- 25,000 images in UK and 3000 in US
AI Beats Radiologists in
Reading Mammograms (3 of
4)
Problem
 High level of information inaccuracy
 Opportunities from new technology
Solutions
 Monitor accuracy and costs
 Evaluate study results
 Collect mammograms for system training
and testing
 Train pattern recognition system
 Compare AI system and human radiologist
findings
Artificial Intelligence Beats
Radiologist in Reading
Mammograms (4 of 4)

Demonstrates how organizational

performance can benefit from
using technology such as artificial
intelligence to facilitate
acquisition and application of
knowledge
Illustrates the ability of AI pattern
recognition systems to achieve
 AI Beats Radiologists in
Reading Mammograms
Monitor
Accuracy &
costs High level of
Evaluate Study inaccuracy
Reports Opportunities from
new technology
Collect
Mammograms for
system training &
testing
Train pattern
recognition
systems AI Mammography
System Optimize
Corporate AI
Read Mammograms Diagnoses
system & human
Classify Improve
radiologist
AI Pattern
mammograms for Treatment
findings
recognition
systems Pathology
What is the Role of Knowledge
Management Systems in Business?
Knowledge management systems
among fastest growing areas of
software investment
Information economy: production and
distribution of information and
knowledge a major source of wealth and
prosperity
Substantial part of a firm’s stock
market value is related to intangible
assets: knowledge, brands, reputations,
and unique business processes
Important Dimensions of
Knowledge (1 of 2)
Data, information, knowledge,
and wisdom
Tacit knowledge and explicit
knowledge
Important dimensions of
knowledge
 Knowledge is a firm asset
 Knowledge has different forms
 Knowledge has a location
Important Dimensions
of Knowledge (2 of 2)
Knowledge-based core competencies
 Key organizational assets
Knowing how to do things effectively and
efficiently in ways others cannot duplicate
is a prime source of profit and competitive
advantage
 Example: Having a unique build-to-order
production system
Organizational learning
 Process in which organizations gain
experience through collection of data,
measurement, trial and error, and
The Knowledge Management Value
Chain (1 of 3)
Knowledge management
 Set of business processes developed in an
organization to create, store, transfer, and
apply knowledge
Knowledge management value chain
 Each stage adds value to raw data and
information as they are transformed into
usable knowledge
▪ Knowledge acquisition
▪ Knowledge storage
▪ Knowledge dissemination
The Knowledge Management
Value Chain (2 of 3)
 Knowledge acquisition
 Documenting tacit and explicit knowledge
▪ Storing documents, reports, presentations,
best practices
▪ Unstructured documents (e.g., e-mails)
▪ Developing online expert networks
 Creating knowledge
 Tracking data from T P S and external sources
 Knowledge storage
 Databases
 Document management systems
 Role of management
The Knowledge Management
Value Chain (3 of 3)
 Knowledge dissemination
 Portals, wikis
 E-mail, instant messaging
 Search engines, collaboration tools
 A deluge of information
▪ Training programs, informal networks, and
shared management experience help
managers focus attention on important
information.
 Knowledge application
 New business practices
 New products and services

The Knowledge Management
Value Chain
Building Organizational and
Management Capital: Collaboration,
Communities of Practice, and Office
Environments
Developing new organizational roles and
responsibilities for the acquisition of
knowledge
Chief knowledge officer executives;
dedicated staff / knowledge managers
Communities of practice (C O P s)
 Informal social networks of professionals
and employees
 Activities include education, online
newsletters, sharing knowledge
 Reduce learning curves of new
employees
Types of Knowledge Management
Systems

 Enterprise-wide knowledge management systems

 General-purpose firm-wide efforts to collect,
store, distribute, and apply digital content and
knowledge
 Knowledge work systems (K W S)
 Specialized systems built for engineers,
scientists, other knowledge workers charged
with discovering and creating new knowledge
 Intelligent techniques
 Diverse group of techniques, such as data
mining, expert systems, machine learning,
used for various goals: discovering knowledge,
distilling knowledge, discovering optimal
Major Types of Knowledge
Management Systems
What Is Artificial
Intelligence?
 Artificial intelligence (AI): a form of intelligent
technique
 Grand vision
 Computer hardware and software systems that
are as “smart” as humans
 So far, this vision has eluded computer
programmers and scientists
 Narrower, more realistic vision
 Systems that take data inputs, process them,
and produce outputs (like all software
programs) and that can perform many complex
tasks that would be difficult or impossible for
Major Types of AI

Expert systems
Machine learning
Neural networks and deep
learning networks
Genetic algorithms
Natural language processing
Computer vision
Robotics
Expert Systems

 Capture tacit knowledge in very specific and

limited domain of human expertise
 Capture knowledge as set of rules
 Typically perform limited tasks
 Diagnosing malfunctioning machine
 Determining whether to grant credit for loan
 Used for discrete, highly structured decision
making
 Knowledge base: Set of hundreds or thousands of
rules
 Inference engine: Strategy used to search
knowledge base
 Forward chaining
Rules in an Expert System
Machine Learning (1 of 2)

Used by neural networks, deep

learning networks, and genetic
algorithms
Different paradigm than expert
systems
Focuses on recognizing patterns in
very large data sets
Contemporary examples
 Facebook ad display
 Netflix recommender system
Machine Learning (2 of 2)

Supervised learning
 System “trained” by providing examples
of desired inputs and outputs identified
by humans in advanced
 One technique used to develop
autonomous vehicles
Unsupervised learning
 Same procedures as used with
supervised learning, but humans do not
provide examples
 “Cat Paper”
Machine Learning
 Separate file
Neural Networks (1 of 3)

Find patterns and relationships in massive

amounts of data too complicated for
humans to analyze
“Learn” patterns by searching for
relationships, building models, and
correcting over and over again
Humans “train” network by feeding it data
inputs for which outputs are known, to help
neural network learn solution by example
from human experts
Used in medicine, science, and business for
problems in pattern classification,
How a Neural Network Works
Neural Networks (2 of 3)
Deep learning neural networks
 More complex, with many layers of
transformations of input data to
produce target output
 Used almost exclusively for pattern
detection on unlabeled data
(unsupervised learning)
 Some believe these come closest
to “grand vision” of AI
A Deep Learning
Network
Neural Networks (3 of 3)

Limitations of neural networks and

machine learning
 Require very large data sets to identify
patterns
 Patterns may not “make sense: or may
be ephemeral
 How system arrived at a particular
solution often cannot be explained
 Most useful for classifying digital assets
into binary categories (yes or no), but
most real-world problems do not have
Genetic Algorithms

Useful for finding optimal solution for

specific problem by examining very large
number of possible solutions for that
problem
Conceptually based on process of evolution
 Search among solution variables by
changing and reorganizing component
parts using processes such as
inheritance, mutation, and selection
Used in optimization problems
(minimization of costs, efficient scheduling,
The Components of a Genetic
Algorithm
Natural Language
Processing
Software that can process
voice or text command using
natural human language
Typically based on machine
learning, including deep
learning
Examples: Google search;
spam filtering systems; text
Computer Vision
Systems
Emulate human visual system to view
and extract information from real-
world images
Examples:
 Facebook’s DeepFace can identify
friends in photos across their
system and the entire web
 Autonomous vehicles can recognize
signs, road markers, people,
animals, and other vehicles with
Robotics
Design, construction, and operation of
movable machines that can substitute for
humans, along with computer systems for
their control, sensory feedback, and
information processing
Generally programmed to perform specific
and detailed actions in limited domains,
e.g. robots spray paint autos, and assemble
certain parts, welding, heavy assembly
movement
Used in dangerous situations like bomb
Intelligent Agents
Work without direct human intervention to
carry out repetitive, predictable tasks
 Deleting junk e-mail
 Finding cheapest airfare
Use limited built-in or learned knowledge
base
 Some are capable of self-adjustment, for
example: Siri
Chatbots
Agent-based modeling applications:
 Model behavior of consumers, stock
Intelligent Agents in P&G’s Supply
Chain Network
What Types of Systems Are Used
for Enterprise-Wide Knowledge
Management?

 Three major types of knowledge in an enterprise

 Structured documents
▪ Reports, presentations
▪ Formal rules
 Semi structured documents
▪ E-mails, videos
 Unstructured, tacit knowledge
 80% of an organization’s business content is semi
structured or unstructured
Enterprise Content Management
Systems

Help capture, store, retrieve,

distribute, preserve documents and
semi structured knowledge
Bring in external sources
 News feeds, research
Tools for communication and
collaboration
 Blogs, wikis, and so on
Key problem: developing taxonomy
Digital asset management systems
An Enterprise Content
Management System
Locating and Sharing
Expertise
Provide online directory of corporate
experts in well-defined knowledge
domains
Search tools enable employees to
find appropriate expert in a company
Social networking and social business
tools for finding knowledge outside
the firm
 Saving
 Tagging
Learning Management Systems (L
M S)

Provide tools for management, delivery,

tracking, and assessment of employee
learning and training
Support multiple modes of learning
 web-based classes, online forums, and
so on
Automates selection and administration of
courses
Assembles and delivers learning content
Measures learning effectiveness
Massively open online courses (M O O C s)
Knowledge Workers and
Knowledge Work
Knowledge workers
 Researchers, designers, architects,
scientists, engineers who create
knowledge for the organization
 Perform key roles critical to organization
and managers who work within
organization
Knowledge work systems
 Systems for knowledge workers to help
create new knowledge and integrate
that knowledge into business
Requirements of Knowledge Work
Systems
 Sufficient computing power for graphics, complex
calculations
 Communications and document management
 Access to external databases
 User-friendly interfaces
 Optimized for tasks to be performed (design
engineering, financial analysis)
Requirements of Knowledge Work
Systems