Intelligent Agents

By
Anand Kumar

1
 Intelligent Software Agents:
An Overview
 Intelligent Agent (IA): Computer program that helps a user
with routine computer tasks
 New Technology
 Other Names
 Software agents
 Wizards
 Knowbots
 Intelligent software robots
 Softbots
 Bots

 Agent: Employing someone to act on your behalf

2
 Definitions of Intelligent Agent
 “Intelligent agents are software entities that carry out some set of
operations on behalf of a user or another program, with some degree of
independence or autonomy and in so doing, employ some knowledge or
representation of the user’s goals or desires.” (“The IBM Agent”)

 An agent is anything that can be viewed as perceiving its environment

through sensors and acting upon that environment through effectors
(Russell and Norvig, 1995, p. 33)

 Autonomous agents are computational systems that inhabit some complex

dynamic environment, sense and act autonomously in this environment
and by doing so realize a set of goals or tasks for which they are designed
(Maes, 1995, p. 108)

3
 More Definitions
 A persistent software entity dedicated to a specific purpose.
“Persistent” distinguishes agents from subroutines; agents have
their own ideas about how to accomplish tasks, e.g., their own
agenda. “Special purpose” distinguishes them from entire
multifunction applications; agents are typically much smaller”
(Smith et al., 1994)

 Intelligent agents continuously perform three functions:

perception of dynamic conditions in the environment; action to
affect conditions in the environment; and reasoning to interpret
perceptions, solve problems, draw inferences and determine
actions (Hayes-Roth, 1995)

4
Introduction
 Little bit history.

 Mid 50’s first ideas about software agents.

• John McCarthy, Oliver G. Selfridge.
 Research on software agents can be split into two main
strands:
 first one about mid 70’s and
 second one at the beginning of 90’s.
 Today software agents are still a hot, diverse topic.
 distributed artificial intelligence, robotics, artificial life,
distributed object computing, human-computer interaction,
intelligent and adaptive interfaces, intelligent search and
filtering, information retrieval, etc.

5
Introduction

 Agents

 Many definitions:

 "Let us define an agent as a persistent software entity dedicated to a specific

purpose. 'Persistent' distinguishes agents from subroutines; agents have their own
ideas about how to accomplish tasks, their own agendas. 'Special purpose'
distinguishes them from entire multifunction applications; agents are typically
much smaller."

 An autonomous agent is a system situated within and a part of

an environment that senses that environment and acts on it,
over time, in pursuit of its own agenda and so as to effect what it
senses in the future.

6
 Introduction
 Categories of agents (just one of them)

Autonomous Agents

Biological Agents Computational Robotic Agents

Agents

Software Agents Artificial Life Agents

Task-specific Agents
Entertainment Agents Viruses

7
Types of Software Agents
 Static (Computer Bound)
 Stationary agents do not roam around the online world
but use embedded knowledge to assist in filtering and
processing the volume of incoming information.
 Agent becomes dormant until the next event arri
 Mobile
 Execute commands while living on a remote server,
reporting back to home page when given task is
accomplished
 Carry out transactions without further input from the
consumer .
8
Why Software Agents
 Managing the information overload
 Decision support
 Repetitive office activity
 Mundane personal activity
 Search and retreival
 Domain experts

9
Properties of mobile software agents
 Programming
 Safety
 Resource usage
 Navigation
 Privacy
 Communication

10
General characteristics of Software
Agents
 Independent Agency
 Agent learning
 Agent cooperation
 Agent reasoning capability
 Rule based approach
 Knowledge based approach

 Learning approach

 Agent interface
11
Technology behind Software Agents
 Components of a software agent
 Owner
 Author

 Lifetime

 Account

 Goal

 Subject description

 Background

12
Technology behind Software Agents
 Components of the agent computing
environment
 User preferences
 Knowledge of what is where

 Cost, resources and time constraints

 Task, process or domain knowledge

13
Technology behind Software Agents
 Telescript
Agent Language
 SAFE-TCL (Tool Command Language)

14
Software agents in action
 Software advisors
 Customer service help desk
 Event monitors
 System and Network Management
 Work-flow assistants and Smart Messaging
 Software Agents and Resource Discovery
 Scale of problem
 Need for cooperative brokerages
 Need for resource planning

15
 Why Intelligent Agents?
Information Overload
 Data doubles annually (in large enterprises (1998))
 Can analyze only about 5%
 Most efforts: discover patterns, not meaning, not what to do
 Reduces decision making capabilities by 50%

 Much caused by the Internet / Web

 How to filter data?
 How to identify relevant sources of data?

 Intelligent agents can assist searching

 Save time: agents decide what is relevant to the user

16
 Internet-based Software Agents
Software Robots or Softbots

Major Categories

 E-mail agents (mailbots)

 Web browsing assisting agents
 Frequently asked questions (FAQ) agents
 Intelligent search (or Indexing) agents
 Internet softbot for finding information
 Network Management and Monitoring

17
Network Management and Monitoring

 Patrol Application Management

 Tabriz
 WatchGuard
 AlertView
 InterAp
 Mercury Center’s Newshound
 Infosage

18
 Electronic Commerce Agents

 Need identification
 Product brokering
 Merchant brokering
 Negotiation
 Purchase and delivery
 Product/service evaluation

19
 Other Agents
 Operating systems agents
 Supply chain management agents
 Spreadsheet agents
 Workflow and administrative management
agents
 Competitive intelligence agents
 Software development agents
 Data mining / Web mining agents
 Monitoring and alerting agents
 Collaboration agents

20
 Operating Systems Agents
Wizards in Microsoft Windows NT Operating Systems
 Add user accounts
 Group management
 Managing file and folder access
 Add printer
 Add/remove programs
 Network client administrator
 Licenses
 Install new modems
 Spreadsheet agents: make software more friendly

21
 Workflow and Administrative
Management Agents

 Ascertain and automate user needs or business processes

 Example - FlowMark

 Software development
 Many routine tasks can be done or supported by agents

22
A Multiagent System for Travel Arrangements
Buyer Sellers

Car Rental
Companies

Car Rental Agents

Airlines

User Agent Airline Agents

Hotels

Hotel Agents

23
 Managerial Issues
 Cost Justification
 Security
 Privacy
 Industrial Intelligence and Ethics
 Other Ethical Issues
 Agent Learning
 Agent Accuracy
 Heightened Expectations
 System Acceptance
 System Technology
 Strategic Information Systems

24
 Conclusions
 Agents can simplify our use of computers

 Agents can provide friendly software assistance

 Agents promise to hide complexity

 Agents perform actions we do not do ourselves

 Agents could enhance human intelligence

 Agents provide support to Net users in handling the information

overload problem

25
 But: Danger!
 Agents are unlike other technological advances

 Agents have some level of intelligence, some form of

 Self-initiated and
 Self-determined goals

 There is the potential for

 Social mischief
 Systems that run amok
 Loss of privacy
 Further alienation of society

26
 Can Eliminate Such Problems
 Develop rules for well-behaving agents

 Determine the accuracy of information collected

 Respect restrictions of other servers

 Do only authorized work

27