Adaptive User Profiling

1
Adaptive User Profiling

• Carolina Bailey
• (cmbail_at_essex.ac.uk)

2
User Profiling Areas

• Information Retrieval
• Personalised Search
• Personalised TV Listings
• Recommendation Systems
• Expert Finding Systems profile matching
• Information Filtering
• Spam Filters
• News Filtering
• E-Learning
• Learner Profiles
• Intelligent Environments
• Intelligent Agents
• Behaviour Prediction

3
User Profiling

• A user profile can be used like a filter on a set
  of data, with various sets of data
• Search Engine results
• Environment variables such as lighting settings
  and temperature settings
• Recipes
• News feeds etc.
• any collection of data items that could be
  personalised
• Any information available about the user can be
  incorporated into the profile
• Likes, dislikes specified or implied
• Various histories e.g. past behaviour,
  purchasing, browsing, TV watching, bookmarks
• Disabilities and/or medical details
• Future data sources

4
User Profiling Steps
5
Building a User Profile

• Various Data Source Applications
• Examples of Data Sources
• HTML file (e.g. bookmarks)
• XML files
• Text files (e.g. rules)
• Various Methods of Processing
• Various Representations of Profiles
• Some Example Personalised Applications and Data
  Sources

6
Data Source Applications

• Search Engine

7
Data Source Applications

• Question Answering
• System

8
Data Source Applications

• Personalised TV

9
Data Source Applications

• Intelligent Environment

10
Data Sources - XML Recipe
Ref Recipe example from http//www.brics.dk/amoe
ller/XML/xml/example.html
11
Data Sources Fuzzy Logic Example
12
Methods of processing data
13
Representations of Profiles
14
A Global, Unified Profile

• Potentially, one single profile could be used
  anywhere, for any application.
• Currently, the common theme in previous research,
  is that there is no common theme!
• Different data storage methods, data processing
  methods and algorithms, representation of
  profiles etc.
• What is the most efficient of these different
  methods and processes?
• Can a user profile from one application be used
  within another application?

15
A Global, Unified Profile
16
A Global, Unified Profile
17
Global Profile - Considerations

• Mapping and categorising items to the Global
  Profile
• E.g. a generic term for temperature, heating,
  radiators etc.
• Extensible way to add new data (and data sources)
  to the profile
• Textual data
• Fuzzy data
• Future data items and sources e.g. SatNav
• Data storage choices
• Main Server
• Distributed
• Transparency of the profile
• Updating and synchronising

18
Security, Privacy and Legal Implications

• The User must be in ultimate control!
• What data should be used in a profile? Purchasing
  history? Criminal record?
• Who and what should be allowed access to a
  profile? The Police? The Government? Could it
  be used against their wishes?
• Fine balance between what is good-intentioned
  personalisation and what is a complete loss of
  privacy
• As people lose more and more control of what
  information is stored about them, their personal
  freedom may feel encroached upon, resulting in a
  strong resistance to further developments towards
  user profiling

19
The End

• To be continued
• cmbail_at_essex.ac.uk