Department of Information Science and Telecommunications

INFSCI 3954: The Adaptive Web

Doctoral Seminar on Cognitive Systems

(Fall 2005, CRN 19141)

[ Formal Data | Syllabus & Schedule | Readings | Course Materials | CourseWeb ]

Syllabus and Schedule

Background

Web systems suffer from an inability to satisfy the heterogeneous needs of many users. For example, Web courses present the same static learning material to students with widely differing knowledge of the subject. Web e-stores offer the same selection of "featured items" to customers with different needs and preferences. Virtual museums on the web offer the same "guided tour'' to visitors with very different goals and interests. Health information sites present the same information to readers with different health problems. A remedy for the negative effects of the traditional "one-size-fits-all'' approach is to develop systems with an ability to adapt their behavior to the goals, tasks, interests, and other features of individual users and groups of users. The Adaptive Web is a relatively young research area. Starting with a few pioneering works on adaptive hypertext in early 1990, it now attracts many researchers from different communities such as hypertext, user modeling, machine learning, natural language generation, information retrieval, intelligent tutoring systems, cognitive science, and Web-based education.

Goal

The goal of the seminar is to get immersed into the world of the Adaptive Web (AW). The attendees will learn about many aspects of AW, get familiar with existing technologies that makes AW possible, and study many examples of modern AW systems. After the end of the seminar the students should be able to develop their own AW systems as well as to apply modern adaptation and personalization technologies in the context of larger projects. Those who will successfully complete the seminar will have a chance to join a small cohort of experts on adaptivity and personalization.

The Web personalization industry is on the rise. In the coming 2-3 years universities, research labs, and companies will need hundreds of experts who understand adaptive systems and personalization to lead various exciting projects. This is your chance to be among the leaders of the new and exciting field

Settings

After a brief introduction into the field provided by the instructor we will switch to a regular work mode. Every week we will read and discuss several research papers or practical adaptive systems and tools. Each classroom session will be devoted to two specific subtopics of the field (see the draft content below) - one particular technology and one particular application or challenge area. For each topic we will have the core presentation, several follow-up presentations, and a post-presentation discussion. The attendees will alternate in leading presentation for each of these topics. Leading a presentation means carefully reading 2-3 review-style papers, preparing the core presentation (about 45 minutes), handling the post-presentation discussion, and summarizing the topic after the session. Those who are not serving as presentation leaders at the session are expected to read one paper on either of the session topics, prepare a summary, and get ready to present an essence of the paper in a follow-up presentation (up to 5 slides). The speakers for each follow-up presentation will be randomly selected from the set of students who prepared a follow-up presentation for this topic.

The overall goal of the seminar is to draft a detailed syllabus and prepare some course material and reading assignment for a hypothetical MS level course on the Adaptive Web. To learn something really well one has to try teaching it. Through our discussion we will try to develop a better structure of the course material. The original set of topics is provided below, but we will have to work together to identify key subtopics and issues within each large topic and to assemble the most relevant readings for each of them.

To complement the "reading part" and to get a real experience in developing adaptive systems the attendees will have to do a small term project. Depending on the level of the programming skills you can either choose to develop a small adaptive information system that uses just 1-2 adaptation technologies, or to develop an adaptive application using one of existing tools such as InterBook (http://www2.sis.pitt.edu/~peterb/InterBook.html ), WBI (http://www.almaden.ibm.com/cs/wbi/ ) or AHA (http://wwwis.win.tue.nl/~debra/ ). (InterBook, for example, will require no programming skills at all).

Topics

Here is the list of topics we will cover during the course

I. Modeling Technologies

1. User models for adaptive hypermedia and adaptive educational systems
2. User profiles for personalized information access
3. Data mining for Web personalization
4. Generic user modeling systems
5. Document modeling

II. Adaptation Technologies

1. Personalized search on the World Wide Web
2. Adaptive focused crawling
3. Adaptive navigation support
4. Collaborative filtering
5. Content-based filtering and recommendation
6. Hybrid Web recommender systems
7. Case-base recommendation
8. Adaptive 3D Web sites
9. Adaptive collaboration support for the Web
10. Adaptive presentation for the Web
11. The rise of the social Web

III. Applications

1. Adaptive systems in health care
2. Adaptive techniques in Web-based education
3. Personalization in e-commerce applications
4. Web-based mobile guides
5. Adaptive news access

IV. Challenges

1. Semantic Web metadata, ontologies, and reasoning for personalized information access on the Web
2. Privacy-enhanced web personalization
3. Open corpus adaptive hypermedia
4. Group recommendation
5. Empirical evaluation of personalized websites

Course Schedule

We have 11 classroom sessions. Each session will have a technology part and an application/challenge part. The exact schedule of session topics can be found on the Course Materials page.

Assignments and Grading

There are four things each student is expected to do:

1. Serving as a presentation leader. Once during the seminar duration each student serves as a presentation leader for the assigned topic. That means reading 2-5 papers on the technology subject (preferably some review papers), preparing a presentation for 40-50 minutes, chairing the session and leading the discussion. Allow at least 2 weeks to prepare your original presentation. After the meeting the leader has to prepare 1-3 page summary of the topics of the meeting that includes his own findings and results of the discussion.
2. Serving as a follow-up presenter. Every week each student reads a (small) paper on the subject of one of the topics of the coming meeting, prepares a 1 page summary and a set of slides, and, if selected, delivers a small 5 minutes follow-up presentation. Printed copies of the summary have to be brought to class to distribute among the attendees.
3. Completing a term project. Each student has to do a term project - implementing a small adaptive system using a programming language (for IS students) or some development tool (for non-programmers). Topics for the project can be discussed with the instructor. It is anticipated that you choose a project closely related to your core research direction so that your work can enhance (rather than harm) the work on your future Thesis.
4. Discovering relevant resources. Each student contributes relevant resources to the pool of resources (relevant papers, known adaptive systems on the Web, known tools, etc). We use a dedicated system for managing contributed resources.

The final grade will be assembled from your performance in:

• Attendance: 5%
• Work as a discussion leader: 30%
• Paper summaries and presentations: 30%
• Resource discovery: 5%
• Final project: 30%

It is assumed that every student will serve as a discussion leader once and will prepare 10 paper summaries (follow-up presentations).

Course Policies

Academic Integrity

Special Considerations

Copyright © 2005 Peter Brusilovsky