1. 1 SEGMENT 10 Intelligent Software Agents and Creativity

2. 2 Intelligent Software Agents: An Overview n Intelligent Agent (IA): Computer program that helps a user with routine computer tasks n New Technology n Other Names –Software agents –Wizards –Knowbots –Intelligent software robots –Softbots –Bots n Agent: Someone employed to act on one’s behalf

3. 3 Definitions of Intelligent Agent n “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”) n 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) n 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)

4. 4 More Definitions n 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) n 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)

5. 5 Intelligence Levels and Power 0: Straight orders 1: User initiated search by key words (search engines) 2: Have user profiles (software agents) 3: Have learning and deductive capabilities (learning or truly intelligent agents)

6. 6 Possible Components of an Agent n Owner n Author n Account n Goal n Subject description n Creation and duration n Background n Intelligent subystem

7. 7 Intelligent Agent Characteristics n Autonomy (empowerment) Agent takes initiative, exercises control over its actions –Goal-oriented –Collaborative –Flexible –Self-starting n Operates in the background –Mobile agents

8. 8 n Communication (interactivity) n Automates repetitive tasks n Reactivity n Proactiveness (persistence) n Temporal continuity n Personality n Mobile agents n Intelligence and learning Single Task

9. 9 Why Intelligent Agents? Information Overload n 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% n Much caused by the Internet/Web –How to filter data –How to identify relevant sources of data n Intelligent agents can assist searching n Save time: agents decide what is relevant to the user

10. 10 n Decision support n Front-line decision support n Repetitive office activity n Mundane personal activity n Search and retrieval n Domain experts Reasons for Intelligent Agent Technology Growth

11. 11 Agent Classification and Types Taxonomic tree to classify autonomous agents (Figure 17.1) Autonomous agents Software agents Computational agents Task-specific agents Artificial life agents Robotic agents Biological agents Entertainmment agentsViruses

12. 12 Application Types n Organizational and personal agents n Private agents vs. public agents n Software (simple) agents and intelligent agents n Mobile agents

13. 13 Classification by Characteristics n Agency n Intelligence n Mobility

14. 14 n Degree of autonomy and authority vested in the agent –Key value of agents –More advanced agents can interact with other entities Agency

15. 15 Intelligence Degree of reasoning and learned behavior

16. 16 Mobility n Degree to which agents travel through the network –Static –Mobile scripts –Mobile with state –Nonmobile agents defined in 2-D –Mobile agents defined in 3-D

17. 17 Classification by Application Area n Assist in workflow and administrative management n Collaborate with other agents and individuals n Support electronic commerce n Support desktop applications n Assist in information access and management n Process mail and messages n Control and manage the network access n Manage systems and networks n Create user interfaces

18. 18 Internet-based Software Agents Software Robots or Softbots Major Categories n E-mail agents (mailbots) n Web browsing assisting agents n Frequently asked questions (FAQ) agents n Intelligent search (or Indexing) agents n Internet softbot for finding information n Network Management and Monitoring

19. 19 n Patrol Application Management n Tabriz n WatchGuard n AlertView n InterAp n Mercury Center’s Newshound n Infosage Network Management and Monitoring

20. 20 Electronic Commerce Agents n Need identification n Product brokering n Merchant brokering n Negotiation n Purchase and delivery n Product/service evaluation

21. 21 Other Agents n Operating systems agents n Supply chain management agents n Spreadsheet agents n Workflow and administrative management agents n Competitive intelligence agents n Software development agents n Data mining / Web mining agents n Monitoring and alerting agents n Collaboration agents

22. 22 Wizards in Microsoft Windows NT Operating Systems n Add user accounts n Group management n Managing file and folder access n Add printer n Add/remove programs n Network client administrator n Licenses n Install new modems n Spreadsheet agents: make software more friendly Operating Systems Agents

23. 23 Workflow and Administrative Management Agents n Ascertain and automate user needs or business processes n Example - FlowMark n Software development –Many routine tasks can be done or supported by agents

24. 24 n One of the most important capabilities of information technology n Can sift through large amounts of information n Challenge: intelligent agents to sift and sort n Categories –Intelligent agents –Query-and-reporting tools –Multidimensional analysis Data Mining

25. 25 Subsets (Etzioni, 1996) n Resource discovery n Information extraction n Generalization Web Mining

26. 26 Monitoring and Alerting: NewsAlert n Monitors data by personalized rules n Automatically delivers alerts to the user’s desktop into personalized newspapers n Organizes alerts by user-specified subject areas n Provides smart tools so users can investigate the context of an alert and communicate findings to others

27. 27 Key Components of NewsAlert n Software agents n Alert objects n Newspaper client

28. 28 Electronic Newspapers n Combine features of a paper newspaper n Familiar format

29. 29 Collaboration by Agents n Lotus Notes/Domino Server: Comprehensive collaborative software n Includes Notes Agents: automates many Notes tasks n Agents operate in the background performing routine tasks n Agents can be created by designers within an application n Agents can either be private or shared n Collaboration: Natural area for agent-to-agent interaction and communication

30. 30 Distributed AI, Multiagents, and Communities of Agents n Software agents must communicate, cooperate and negotiate with each other n Refine requests and queries through evolving dialogue n Intelligent agents work together in multiple agent systems n Agents can communicate, cooperate and/or negotiate n Easy to build agents with small specialized knowledge n But complex tasks require much knowledge n Agents need to share their knowledge

31. 31 A Multiagent System for Travel Arrangements

32. 32 n Agents control a telecommunications network n Can enter into agreements with other computers that control other networks about routing packets more efficiently n Agent in a blackboard architecture Routing in Telecommunication Networks

33. 33 n Personal digital assistants (PDA) n Shared (global) databases n Agents (softbots) travel out on the Internet and collect information from shared databases n Traffic control n Coordination of vehicular traffic n Air traffic control n The University of Massachusetts CIG Searchbots n Software agents make decisions based on communication and agreements with other agents n Soon: Agents coordinating sellers and buyers More Multiple Agents

34. 34 Topics in Multiagent Systems n Negotiation in electronic commerce n Coordination n The nature of the agents n Learning agents n Cooperation and collaboration n Communities of agents

35. 35 n Data monitoring n Data gathering n Modeling n Domain managing n Preference learning DSS Agents

36. 36 n Cost Justification n Security n Privacy n Industrial Intelligence and Ethics n Other Ethical Issues n Agent Learning n Agent Accuracy n Heightened Expectations n System Acceptance n System Technology n Strategic Information Systems Managerial Issues

37. 37 n Agents can simplify our use of computers n Agents can provide friendly software assistance n Agents promise to hide complexity n Agents perform actions we do not do ourselves n Agents could enhance human intelligence n Agents provide support to Net users in handling the information overload problem Conclusions

38. 38 n Agents are unlike other technological advances n Agents have some level of intelligence, some form of –Self-initiated and –Self-determined goals n There is the potential for –Social mischief –Systems that run amok –Loss of privacy –Further alienation of society But: Danger!

39. 39 Can Eliminate Such Problems n Develop rules for well-behaving agents n Determine the accuracy of information collected n Respect restrictions of other servers n Do only authorized work