Multi-topic information filtering with a single user profile

Nikolaos Nanas, Victoria Uren, Anne de Roeck, John Domingue

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In Information Filtering (IF) a user may be interested in several topics in parallel. But IF systems have been built on representational models derived from Information Retrieval and Text Categorization, which assume independence between terms. The linearity of these models results in user profiles that can only represent one topic of interest. We present a methodology that takes into account term dependencies to construct a single profile representation for multiple topics, in the form of a hierarchical term network. We also introduce a series of non-linear functions for evaluating documents against the profile. Initial experiments produced positive results.
Original languageEnglish
Title of host publicationMethods and applications of artificial intelligence
Subtitle of host publicationthird Hellenic conference on AI, SETN 2004, Samos, Greece, May 5-8, 2004. Proceedings
EditorsGeorge A. Vouros, Themistoklis Panayiotopoulos
Place of PublicationBerlin (DE)
PublisherSpringer
Pages400-409
Number of pages10
ISBN (Electronic)978-3-540-24674-9
ISBN (Print)978-3-540-21937-8
DOIs
Publication statusPublished - 2004
Event3rd Hellenic Conference on Artificial Intelligence - Samos, Greece
Duration: 5 May 20048 May 2004

Publication series

NameLecture notes in computer science
PublisherSpringer
Volume3025
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference3rd Hellenic Conference on Artificial Intelligence
Abbreviated titleSETN 2004
CountryGreece
CitySamos
Period5/05/048/05/04

Fingerprint

Information filtering
Information retrieval
Experiments

Cite this

Nanas, N., Uren, V., de Roeck, A., & Domingue, J. (2004). Multi-topic information filtering with a single user profile. In G. A. Vouros, & T. Panayiotopoulos (Eds.), Methods and applications of artificial intelligence: third Hellenic conference on AI, SETN 2004, Samos, Greece, May 5-8, 2004. Proceedings (pp. 400-409). (Lecture notes in computer science; Vol. 3025). Berlin (DE): Springer. https://doi.org/10.1007/978-3-540-24674-9_42
Nanas, Nikolaos ; Uren, Victoria ; de Roeck, Anne ; Domingue, John. / Multi-topic information filtering with a single user profile. Methods and applications of artificial intelligence: third Hellenic conference on AI, SETN 2004, Samos, Greece, May 5-8, 2004. Proceedings. editor / George A. Vouros ; Themistoklis Panayiotopoulos. Berlin (DE) : Springer, 2004. pp. 400-409 (Lecture notes in computer science).
@inproceedings{e06e103696aa4ae79ff27b6fb457e387,
title = "Multi-topic information filtering with a single user profile",
abstract = "In Information Filtering (IF) a user may be interested in several topics in parallel. But IF systems have been built on representational models derived from Information Retrieval and Text Categorization, which assume independence between terms. The linearity of these models results in user profiles that can only represent one topic of interest. We present a methodology that takes into account term dependencies to construct a single profile representation for multiple topics, in the form of a hierarchical term network. We also introduce a series of non-linear functions for evaluating documents against the profile. Initial experiments produced positive results.",
author = "Nikolaos Nanas and Victoria Uren and {de Roeck}, Anne and John Domingue",
year = "2004",
doi = "10.1007/978-3-540-24674-9_42",
language = "English",
isbn = "978-3-540-21937-8",
series = "Lecture notes in computer science",
publisher = "Springer",
pages = "400--409",
editor = "Vouros, {George A.} and Themistoklis Panayiotopoulos",
booktitle = "Methods and applications of artificial intelligence",
address = "Germany",

}

Nanas, N, Uren, V, de Roeck, A & Domingue, J 2004, Multi-topic information filtering with a single user profile. in GA Vouros & T Panayiotopoulos (eds), Methods and applications of artificial intelligence: third Hellenic conference on AI, SETN 2004, Samos, Greece, May 5-8, 2004. Proceedings. Lecture notes in computer science, vol. 3025, Springer, Berlin (DE), pp. 400-409, 3rd Hellenic Conference on Artificial Intelligence, Samos, Greece, 5/05/04. https://doi.org/10.1007/978-3-540-24674-9_42

Multi-topic information filtering with a single user profile. / Nanas, Nikolaos; Uren, Victoria; de Roeck, Anne; Domingue, John.

Methods and applications of artificial intelligence: third Hellenic conference on AI, SETN 2004, Samos, Greece, May 5-8, 2004. Proceedings. ed. / George A. Vouros; Themistoklis Panayiotopoulos. Berlin (DE) : Springer, 2004. p. 400-409 (Lecture notes in computer science; Vol. 3025).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Multi-topic information filtering with a single user profile

AU - Nanas, Nikolaos

AU - Uren, Victoria

AU - de Roeck, Anne

AU - Domingue, John

PY - 2004

Y1 - 2004

N2 - In Information Filtering (IF) a user may be interested in several topics in parallel. But IF systems have been built on representational models derived from Information Retrieval and Text Categorization, which assume independence between terms. The linearity of these models results in user profiles that can only represent one topic of interest. We present a methodology that takes into account term dependencies to construct a single profile representation for multiple topics, in the form of a hierarchical term network. We also introduce a series of non-linear functions for evaluating documents against the profile. Initial experiments produced positive results.

AB - In Information Filtering (IF) a user may be interested in several topics in parallel. But IF systems have been built on representational models derived from Information Retrieval and Text Categorization, which assume independence between terms. The linearity of these models results in user profiles that can only represent one topic of interest. We present a methodology that takes into account term dependencies to construct a single profile representation for multiple topics, in the form of a hierarchical term network. We also introduce a series of non-linear functions for evaluating documents against the profile. Initial experiments produced positive results.

UR - http://link.springer.com/chapter/10.1007%2F978-3-540-24674-9_42

U2 - 10.1007/978-3-540-24674-9_42

DO - 10.1007/978-3-540-24674-9_42

M3 - Conference contribution

SN - 978-3-540-21937-8

T3 - Lecture notes in computer science

SP - 400

EP - 409

BT - Methods and applications of artificial intelligence

A2 - Vouros, George A.

A2 - Panayiotopoulos, Themistoklis

PB - Springer

CY - Berlin (DE)

ER -

Nanas N, Uren V, de Roeck A, Domingue J. Multi-topic information filtering with a single user profile. In Vouros GA, Panayiotopoulos T, editors, Methods and applications of artificial intelligence: third Hellenic conference on AI, SETN 2004, Samos, Greece, May 5-8, 2004. Proceedings. Berlin (DE): Springer. 2004. p. 400-409. (Lecture notes in computer science). https://doi.org/10.1007/978-3-540-24674-9_42