Experience with Rule Induction and k-Nearest Neighbor Methods for Interface Agents that Learn

T R Payne, Peter Edwards, Claire Louise Green

Research output: Contribution to journalArticle

12 Citations (Scopus)

Abstract

Interface agents are being developed to assist users with a variety of tasks. To perform effectively, such agents need knowledge of user preferences. An agent architecture has been developed which observes a user performing tasks, and identifies features which can be used as training data by a learning algorithm. Using the learned profile, an agent can give advice to the user on dealing with new situations. The architecture has been applied to two different information filtering domains: classifying incoming mail messages (Magi) and identifying interesting USENet news articles (UNA). This paper describes the architecture and examines the results of experimentation with different learning algorithms and different feature extraction strategies within
these domains.
Original languageEnglish
Pages (from-to)329-335
JournalIEEE Transactions on Knowledge and Data Engineering
Volume9
Issue number2
Publication statusPublished - 1997

Fingerprint

Learning algorithms
Information filtering
Feature extraction

Keywords

  • machine learning
  • software agent
  • rule induction
  • information filtering
  • instance-based learning

Cite this

Experience with Rule Induction and k-Nearest Neighbor Methods for Interface Agents that Learn. / Payne, T R ; Edwards, Peter; Green, Claire Louise.

In: IEEE Transactions on Knowledge and Data Engineering, Vol. 9, No. 2, 1997, p. 329-335.

Research output: Contribution to journalArticle

@article{415f9a19371a4800ac6c2d9e5d99deca,
title = "Experience with Rule Induction and k-Nearest Neighbor Methods for Interface Agents that Learn",
abstract = "Interface agents are being developed to assist users with a variety of tasks. To perform effectively, such agents need knowledge of user preferences. An agent architecture has been developed which observes a user performing tasks, and identifies features which can be used as training data by a learning algorithm. Using the learned profile, an agent can give advice to the user on dealing with new situations. The architecture has been applied to two different information filtering domains: classifying incoming mail messages (Magi) and identifying interesting USENet news articles (UNA). This paper describes the architecture and examines the results of experimentation with different learning algorithms and different feature extraction strategies withinthese domains.",
keywords = "machine learning, software agent, rule induction, information filtering, instance-based learning",
author = "Payne, {T R} and Peter Edwards and Green, {Claire Louise}",
year = "1997",
language = "English",
volume = "9",
pages = "329--335",
journal = "IEEE Transactions on Knowledge and Data Engineering",
issn = "1041-4347",
publisher = "IEEE COMPUTER SOC",
number = "2",

}

TY - JOUR

T1 - Experience with Rule Induction and k-Nearest Neighbor Methods for Interface Agents that Learn

AU - Payne, T R

AU - Edwards, Peter

AU - Green, Claire Louise

PY - 1997

Y1 - 1997

N2 - Interface agents are being developed to assist users with a variety of tasks. To perform effectively, such agents need knowledge of user preferences. An agent architecture has been developed which observes a user performing tasks, and identifies features which can be used as training data by a learning algorithm. Using the learned profile, an agent can give advice to the user on dealing with new situations. The architecture has been applied to two different information filtering domains: classifying incoming mail messages (Magi) and identifying interesting USENet news articles (UNA). This paper describes the architecture and examines the results of experimentation with different learning algorithms and different feature extraction strategies withinthese domains.

AB - Interface agents are being developed to assist users with a variety of tasks. To perform effectively, such agents need knowledge of user preferences. An agent architecture has been developed which observes a user performing tasks, and identifies features which can be used as training data by a learning algorithm. Using the learned profile, an agent can give advice to the user on dealing with new situations. The architecture has been applied to two different information filtering domains: classifying incoming mail messages (Magi) and identifying interesting USENet news articles (UNA). This paper describes the architecture and examines the results of experimentation with different learning algorithms and different feature extraction strategies withinthese domains.

KW - machine learning

KW - software agent

KW - rule induction

KW - information filtering

KW - instance-based learning

M3 - Article

VL - 9

SP - 329

EP - 335

JO - IEEE Transactions on Knowledge and Data Engineering

JF - IEEE Transactions on Knowledge and Data Engineering

SN - 1041-4347

IS - 2

ER -