Open AccessAn adaptive user profile for filtering news based on a user interest hierarchy
Author(s) -
Singh Sarabdeep,
Author Michael Shepherd Corresponding,
Duffy Jack,
Watters Carolyn
Publication year - 2006
Publication title -
proceedings of the american society for information science and technology
Language(s) - English
Resource type - Journals
eISSN - 1550-8390
pISSN - 0044-7870
DOI - 10.1002/meet.1450430154
Subject(s) - session (web analytics) , computer science , hierarchy , user modeling , ranking (information retrieval) , user profile , recall , computer user satisfaction , human–computer interaction , information retrieval , user interface , world wide web , programming language , psychology , economics , market economy , cognitive psychology
A prototype system for the filtering and ranking of news items has been developed and a pilot test has been conducted. The user's interests are modeled by a user interest hierarchy based on explicit user feedback with adaptive learning after each session. The system learned very quickly, reaching normalized recall values of over 0.9 within three sessions. When the user's interests “drifted”, the system adapted but the speed with which it adapted seemed dependent on the amount of feedback provided by the user.