html5-img
1 / 17

Pertemuan 21 Software Agents for E-Commerce

Pertemuan 21 Software Agents for E-Commerce. Matakuliah : M0284/Teknologi & Infrastruktur E-Business Tahun : 2005 Versi : <<versi/revisi>>. Learning Objectives. Describe what software agents are Differentiate between various classes of software agents

thor
Download Presentation

Pertemuan 21 Software Agents for E-Commerce

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Pertemuan 21Software Agents for E-Commerce Matakuliah : M0284/Teknologi & Infrastruktur E-Business Tahun : 2005 Versi : <<versi/revisi>>

  2. Learning Objectives • Describe what software agents are • Differentiate between various classes of software agents • Understand the use of artificial intelligence and statistical reasoning • Describe the range of agents available to assist in the buying process • Identify various activities in e-commerce where software agents can be used

  3. Overview • What are software agents? • Logic of agent behavior • Types of agents • Information agents • E-Commerce agents • Mobile agents

  4. Software entities Autonomy/agency - without detailed commands Purposeful - goal-driven Reactive - react to changes in environment. Exhibit intelligence Social and Mobility skill - travel around and interact with other agents What are software agents?

  5. What are Software Agents?

  6. Logic of Agent Behavior • Symbolic Reasoning if <condition> then <action> • BeyondMail from Banyan 1) IF <event='mail_receipt'> AND < email_sender=’CEO’> THEN <save_in_folder=’Urgent’> 2) IF <save_in_folder NOT Empty> THEN <notify>

  7. Logic of Agent Behavior • Statistical Reasoning • Market Segmentation • Clustering according to some characteristics such a buying behavior, demographic data • Also called Collaborative filtering • Used by Amazon.com to predict books that might prove to be your favorite

  8. Logic of Agent Behavior • Multi-attribute utility theory used to rank-order different choices such as items to buy Utility is related to various quality, price and delivery attributes Utility numbers are calculated for various choicesVarious formulas used: U(x)= log( x+ b) U(x)= a + bx + cx2 U(x) = (1/k) (1- e –kx) where U is the utility and x is the measure of the attribute. In the case of an automobile, x could be price, quality or fuel economy.

  9. Logic of Agent Behavior • Constraint Satisfaction Approach • A way to prune a large set of choices • Hard and Soft constraints • Options/ choices that violate hard constraints are removed • Options left are evaluated in terms of how far soft constraints violated

  10. Logic of Agent Behavior • Auction Protocols • English auction price start low and move up • Dutch auction price start high and move low • Sealed-bid auction offers in sealed envelopes

  11. Logic of Agent Behavior Auction Engines used in e-business

  12. Types of Software Agents • Information Agents • E-Commerce Agents • Mobile Agents

  13. Information Agents • Information Search Agents search engines • Information filtering agents search few specific web site and retrieve information relevant to a user

  14. Information Agents Logic of filtering agents

  15. Information Delivery Agents Pull versus Push (scheduled pull). In push, the client-based software periodically contacting the server for recent news Information Agents

  16. Information Agents • Information Notification Agents • message arrives by email

  17. Information Agents • Information Reconnaissance Agents • Letizia at MIT brings to attention to users pages of interest that are only a few links away from the current page • The system builds up a interest profile of the user and searches neighboring pages of interest

More Related