Скачать презентацию Information Architecture Personalized User Experiences CHI 2003 Скачать презентацию Information Architecture Personalized User Experiences CHI 2003

3587817075fb40ea099caf71cc794266.ppt

  • Количество слайдов: 19

Information Architecture & Personalized User Experiences CHI 2003 Workshop on Designing Personalized User Experiences Information Architecture & Personalized User Experiences CHI 2003 Workshop on Designing Personalized User Experiences for e. Commerce April 6, 2003 Keith Instone keith@instone. org http: //keith. instone. org/ © 2003 Keith Instone

Who am I, where am I from? • Information Architect, ibm. com User Experience Who am I, where am I from? • Information Architect, ibm. com User Experience team • IA Consultant, Argus Associates • Retiree: HCI VL, usableweb. com, CHIWEB, CHI volunteer • Computer Science HCI Hypertext Web usability IA • Ohio

“Traditional” Information Architecture • Combination of organization, labeling and navigation schemes within an information “Traditional” Information Architecture • Combination of organization, labeling and navigation schemes within an information system • Structural design of an information space to facilitate task completion and access to content • Art and science of structuring and classifying web sites/intranets to help people find and manage information • Emerging discipline and community of practice focused on bringing principles of design and architecture to the digital landscape

IA & Personalization • The information architect focuses on how things are structured within IA & Personalization • The information architect focuses on how things are structured within the user experience – Looks “up” to the user interface – how the navigation and page layout convey the structure – Looks “down” to the content management to make sure it can enable to right UE User Interface Layer Information Architecture Layer Content Management Layer Library science applied to personalization

IA Framework for Personalization Business rules map user profiles to content tagging to create IA Framework for Personalization Business rules map user profiles to content tagging to create individual user experiences Information Architecture Layer

Example of the Framework: Profile Layer • “Fashion” rules – If user: style = Example of the Framework: Profile Layer • “Fashion” rules – If user: style = “hip” then You_may_also_like = (content: color = “hot pink” or “black”) and (content: watch: type = “imported”) and (content: outerwear: fabric = “gore tex”) • Do not want to tag everything “hip”

Our business rule personalizes these products for users who are “hip” Our business rule personalizes these products for users who are “hip”

Examples of the Framework: Profile Layer • One-time rules – If user: past purchases Examples of the Framework: Profile Layer • One-time rules – If user: past purchases = “Ponte Di Roma Jumper” then show content: promotion = “PDRJ last chance sale” • Quick rule, “on sale right now”

Profile rule: One time promotions Our business rule personalizes this promo to take advantage Profile rule: One time promotions Our business rule personalizes this promo to take advantage of an immediate opportunity Last chance to buy more Ponte Di Roma Jumpers! They have been discontinued but we still have a few in stock. As a past purchaser of these soft knit dresses, you have the first chance to buy these before they are all gone. Click here to see what is left…

Examples of the Framework: Vocabulary Layer • Show content: promotion: product that matches user: Examples of the Framework: Vocabulary Layer • Show content: promotion: product that matches user: past purchases – Requires content: promotion: product to be same vocabulary as user: past purchases – Works for all product promotions • Show You_may_also_like such that user: style = content: style – Requires content to be tagged “hip” to match user profile value of “hip” – Works for all styles

Research Issue: Where am I vs. Who am I? • Lost in hyperspace – Research Issue: Where am I vs. Who am I? • Lost in hyperspace – “obsession” with telling users where they are • Personalization is more about “who am I” – what do you know about me, how is it being used

Where am I? Site > Store > Department > Section > Page Site Store Where am I? Site > Store > Department > Section > Page Site Store Department Section Page Section > Page

If I buy something for my wife, will it mess everything up? Who am If I buy something for my wife, will it mess everything up? Who am I? Am I logged in? Are these all of the shirts or are you filtering based on my profile? Does this mean you think I like pink? Does this order mean anything – like that I hate short sleeved shirts? Is this a special offer just for me?

The Navigation Metaphor The Navigation Metaphor

The Subscription Metaphor The Subscription Metaphor

The Personalized Navigation Metaphor? X Blah blah Personalized Content Pers. Nav. Where am I? The Personalized Navigation Metaphor? X Blah blah Personalized Content Pers. Nav. Where am I? Local Navigation Global Navigation Who am I?

Questions • How do we tell users who they are? • Do we throw Questions • How do we tell users who they are? • Do we throw away “where am I” conventions and the navigation metaphor? • How do we mix the metaphors? How do users perceive the differences?

UI / Content Framework for Personalization Who personalizes? What is personalized? User explicitly personalizes UI / Content Framework for Personalization Who personalizes? What is personalized? User explicitly personalizes the user interface User explicitly personalizes the content Customize “skin” for the UI Subscribe to channel System implicitly personalizes the user interface System implicitly personalizes the content Browser detection Recommendations based on prior purchases

User Involvement Framework for Personalization Who sets profile? Which profile is set? User explicitly User Involvement Framework for Personalization Who sets profile? Which profile is set? User explicitly sets user profile User explicitly sets content profile Fill out form with zip code Brand popularity assigned System implicitly sets user System implicitly profile content profile Product interest based on browsing pattern Products purchased at same time