Automated Web Page Synthesis for Adaptive Web Systems

1
Automated Web Page Synthesis for Adaptive Web
Systems
Hossein Sadat and Ali A. Ghorbani (t15h7,
ghorbani_at_unb.ca)

• Introduction
• Adaptive Web Systems
• Producing response based on
• User behavior and preferences
• Environment
• Technology
• Dynamic Page Generation
• Different degrees of dynamism exist
• Dynamic page generation is a requirement for
  adaptation
• Adaptive Hypermedia Systems (AHS) can be defined
  by
• User Model
• Domain Model
• Adaptation Model
• We would like to formalize dynamic page
  generation based on the three models of AHS
• Page Synthesis putting information fragments
  together to produce a page

• Page Synthesis Process
• Consists of
• Request analysis and concept extraction
• Information fragments (components) extraction
• Adaptation rules/programs execution and final
  components selection
• Page template and presentation attributes
  selection
• Final page generation

• The Proposed Architecture
• Application Description
• Component Repository
• All the information fragments are stored in this
  repository
• The fragments are stored with a format that
  specifies the content of the fragment and its
  relationship to the domain concepts
• User Model Structure
• Application dependent? should be defined for each
  application
• Two parts demographic attributes and
  application-specific attributes
• Adaptation Description
• Describes when and how the application adapts
• Can be described by a program
• These programs then can be executed by the
  Interpreter
• Domain Model Description
• Describes the concept hierarchy and other
  relationships between conceptual entities in the
  application

• The Proposed Architecture
• Web Server
• A conventional Web server, responsible for
  handling HTTP requests
• Filter
• A processing unit preprocessing the requests and
  post-processing the responses
• Front controller of the system
• Converts an HTTP request to a synthesis request
• Adaptation Server
• Responsible for producing a page for a request
• Creates, manages and releases processes for
  clients requests
• Session Process

• Related Work
• Index Page Synthesis
• Index page a page consisting of links to a set
  of pages that cover a particular topic
• The synthesis of an index page is divided into
  several sub-problems, such as determining the
  contents, their order, the title of the page,
  etc.
• Only the first sub-problem, that is determining
  the contents of the page, has been investigated
• SETA
• A prototype toolkit for building adaptive Web
  stores
• Dynamically selects the content of a page based
  on the users interests and familiarity with the
  products
• Sorts the items of a product class based on the
  users preferences
• SeAN
• An adaptive system for personalized access to
  news
• Three goals
• Selecting news topics relevant to the user
• Presenting appropriate level of details for the
  news
• Providing advertisement most relevant to the user
  and current page
• Each piece of news is considered as a composite
  entity having several attributes defining its
  components
• Title, abstract, full text, author, pictures,
  video can be attributes
• Based on this representation, different levels of
  details can be used for a piece of news based on
  the user model

• Conclusion
• Page Synthesis was defined as a systematic
  procedure for dynamically generating Web pages in
  adaptive hypermedia systems
• Page synthesis consists of several steps request
  analysis, component selection, adaptation program
  execution and components refinement, template
  selection, and final page rendering
• Java Servlet technology is used for Web front
  end, though the proposed architecture imposes no
  restriction
• The proposed architecture is domain-independent
  and may be used for a variety of Web systems

• Future Work
• Future work includes
• Integration of the architecture with available
  platforms, such as J2EE and .NET
• Simplifying the authoring tasks
• Content authoring
• Investigating the best trade-off between the
  degree of programmability of the adaptation
  server and the complexity of authoring tasks
• Find out if rules are powerful enough to describe
  an adaptation model for this architecture
• Providing an evaluation framework for this
  architecture