Transcript 355_Intelligent Agent_YousefiK - cct355-f12

Intelligent Agent
By: Kian Yousefi
Intelligent Agent
• In artificial intelligence, an intelligent
agent (IA) is which observes and acts
upon an environment (i.e. it is an
agent) and directs its activity towards
achieving goals (i.e. it is rational).
Intelligent agents may also learn or
use knowledge to achieve their goals.
Intelligent Agent
• An intelligent agent is software that assists
you, or acts on your behalf, in performing
repetitive computer related tasks
– i.e. clippie (the annoying Microsoft paperclip)
– i.e. Amazon’s recommendation system
• The four main types of intelligent agents
are:
–
–
–
–
buyer agents or shopping bots,
user or personal agents,
monitoring and surveillance agents,
data mining agents
Buyer Agent or Shopping
Bots (Information Agents)
• Is an intelligent agent on a web site
that helps the customer, find products
and services they want.
• Shopping bots work very efficiently
when it comes to searching for
commodity products.
Types of Buyer agents
• There are three types of buyer
agents for online shopping
environments:
– Profile filtering
– Psychographic filtering
– Adaptive filtering
• These agents use different methods
to find recommendations for the
customers.
Types of Buyer agents
– Profile filtering requires that you choose terms
or enter keywords
• Provides a more personal picture of you and your
preferences
– Psychographic filtering anticipates your
preferences on the basis of the answers you
give to questionnaire
– Adaptive filtering asks you to rate products or
situations and also monitors your actions over
time to find out what you like and dislike
Buyer agents or shopping
bots (example)
• CNET's shopping bot focuses on
computers and other types of consumer
electronics, including wireless devices.
One of the earliest shopping bots, Price
Watch has allowed shoppers to search for
computers and computer parts since 1995.
Buyer agents or shopping
bots (example)
• Amazon’s agent uses collaborative
filtering, which consists of matching
each customer with a group of users
who have similar tastes and presents
them with similar products.
Buyer agents or shopping
bots (example)
• BookFinder.com searches Amazon,
Antiqbook, Barnes and Noble,
Bibliofind, and other online book
retailers. A combined database of
roughly 15 million books to find the
cheapest book prices for shoppers.
Monitoring and Surveillance
Agents (Predictive Agents)
• Are intelligent agents that observe and
report on equipment
• These agents look for a change in
conditions
– Often used to monitor manufacturing
equipment, complex computer networks,
inventory, planning, etc.
– May also be used to monitor competition,
Internet sites, discussion groups, mailing lists,
products, and auction sites
Monitoring agents (Example)
• NASA's Laboratory has an
agent that monitors inventory,
planning, and schedule
equipment ordering. This
assures that everything is
ordered on time and keeps
cost low.
Monitoring agents (Example)
• Allstate Insurance has a network with
thousands of computers. The
company uses a network monitoring
agent that watches its huge networks
24 hours a day. Every five seconds,
the agent measures 1200 data points
and can predict a system crash 45
minutes before it happens.
Data Mining Agents
• These intelligent agents operate in a
data warehouse and work to discover
information
• A common type of data mining is
classification
– which finds patterns of information
and categorizes items into those
classes
Data Mining Agents
• Data mining agents
– Go through a lot of information for
relevant information
– Detect a major shift in the trend or a key
indicator
– Detect the presence of new information
and alert you
User or Personal Agents
•
Intelligent agents that take action on your
behalf. For example,
– Check your email, sort it and alert you of
important emails
– Play computer games as your opponent or
patrol game areas for you
– Assemble customized news reports
– Find information on subjects
– Fill out web forms
– Scan web pages and highlight important text
Features of Personal Agents
• Adaptability: Personal agents must be
able to learn the user’s behavior, the users
habits and preferences, and be able to
adapt to them,
• Information Sharing: often user’s
personal agents will find it beneficial to
share information with each other rather
than simply work alone.
• Agent Collaboration: Personal
agents may need to collaborate with
each other to perform tasks for their
users.
• Information Privacy: Users must trust
their personal agents. These agents have
access to the users personal information
since in a lot of cases they are performing
actions on behalf of the user.
•
•
•
•
Haag & Cummings & McCubbrey & Pinsonneault & Donovan
(2006). Management Information Systems Third Canadian
Ed. McGraw-Hill Ryerson
http://eil.utoronto.ca/profiles/rune/node6.html
http://searchcio.techtarget.com/definition/intelligent-agent
http://www.cs.cmu.edu/~softagents/
Time’s Up!
• About your speaker:
– Name: Kian Yousefi
– Prepared: Michael Jones CCT355
– Quick bio: 3rd year digital enterprise
management student.