Efficient Topic Detection Using an Adaptive Neural Network Architecture

Meriem Manai; Sadok Ben Yahia

doi:10.1007/978-3-031-51664-1_10

Efficient Topic Detection Using an Adaptive Neural Network Architecture

Meriem Manai^*
, Sadok Ben Yahia

^*Kontaktforfatter

Publikation: Kapitel i bog/rapport/konference-proceeding › Konferencebidrag i proceedings › Forskning › peer review

Abstract

Topic detection is the process of identifying the underlying themes or topics present in a set of documents. It has become more critical due to the increase of information electronically available and the necessity to process and filter it. In this respect, we introduce a new approach to detecting topics called ClusART. Thus, we created a three-phase approach: a first phase during which lexical preprocessing was conducted. The second phase pays heed to the construction and the generation of vectors representing the documents carried out. In the topic detection phase, the FuzzyART algorithm is used for the training phase, and a classifier based on ParagraphVector is used for the test phase. The comparative study of our approach on the 20 Newsgroups dataset showed that our method could detect almost relevant topics.

Originalsprog	Engelsk
Titel	Advances in Information Systems, Artificial Intelligence and Knowledge Management - 6th International Conference on Information and Knowledge Systems, ICIKS 2023, Proceedings
Redaktører	Inès Saad, Camille Rosenthal-Sabroux, Faiez Gargouri, Salem Chakhar, Nigel Williams, Ella Haig
Forlag	Springer Science+Business Media
Publikationsdato	2024
Sider	145-157
ISBN (Trykt)	9783031516634
DOI	https://doi.org/10.1007/978-3-031-51664-1_10
Status	Udgivet - 2024
Udgivet eksternt	Ja
Begivenhed	6th International Conference on Information and Knowledge Systems, ICIKS 2023 - Portsmouth, Storbritannien Varighed: 22. jun. 2023 → 23. jun. 2023

Konference

Konference	6th International Conference on Information and Knowledge Systems, ICIKS 2023
Land/Område	Storbritannien
By	Portsmouth
Periode	22/06/2023 → 23/06/2023

Navn	Lecture Notes in Business Information Processing
Vol/bind	486
ISSN	1865-1348

Bibliografisk note

Publisher Copyright:
© 2024, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Dokumenter og links

10.1007/978-3-031-51664-1_10

SDU link

Tjek adgangen til materialet i Syddansk Universitetsbiblioteks søgemaskine

Citationsformater

Manai, M., & Ben Yahia, S. (2024). Efficient Topic Detection Using an Adaptive Neural Network Architecture. I I. Saad, C. Rosenthal-Sabroux, F. Gargouri, S. Chakhar, N. Williams, & E. Haig (red.), Advances in Information Systems, Artificial Intelligence and Knowledge Management - 6th International Conference on Information and Knowledge Systems, ICIKS 2023, Proceedings (s. 145-157). Springer Science+Business Media. https://doi.org/10.1007/978-3-031-51664-1_10

Manai, Meriem ; Ben Yahia, Sadok. / Efficient Topic Detection Using an Adaptive Neural Network Architecture. Advances in Information Systems, Artificial Intelligence and Knowledge Management - 6th International Conference on Information and Knowledge Systems, ICIKS 2023, Proceedings. red. / Inès Saad ; Camille Rosenthal-Sabroux ; Faiez Gargouri ; Salem Chakhar ; Nigel Williams ; Ella Haig. Springer Science+Business Media, 2024. s. 145-157 (Lecture Notes in Business Information Processing, Bind 486).

@inproceedings{b57c3a3633444ea88506fa8a49a2609c,

title = "Efficient Topic Detection Using an Adaptive Neural Network Architecture",

abstract = "Topic detection is the process of identifying the underlying themes or topics present in a set of documents. It has become more critical due to the increase of information electronically available and the necessity to process and filter it. In this respect, we introduce a new approach to detecting topics called ClusART. Thus, we created a three-phase approach: a first phase during which lexical preprocessing was conducted. The second phase pays heed to the construction and the generation of vectors representing the documents carried out. In the topic detection phase, the FuzzyART algorithm is used for the training phase, and a classifier based on ParagraphVector is used for the test phase. The comparative study of our approach on the 20 Newsgroups dataset showed that our method could detect almost relevant topics.",

keywords = "FuzzyART, PragraphVector classifier, Text clustering, Topic detection",

author = "Meriem Manai and \{Ben Yahia\}, Sadok",

note = ".; 6th International Conference on Information and Knowledge Systems, ICIKS 2023 ; Conference date: 22-06-2023 Through 23-06-2023",

year = "2024",

doi = "10.1007/978-3-031-51664-1\_10",

language = "English",

isbn = "9783031516634",

series = "Lecture Notes in Business Information Processing",

publisher = "Springer Science+Business Media",

pages = "145--157",

editor = "In{\`e}s Saad and Camille Rosenthal-Sabroux and Faiez Gargouri and Salem Chakhar and Nigel Williams and Ella Haig",

booktitle = "Advances in Information Systems, Artificial Intelligence and Knowledge Management - 6th International Conference on Information and Knowledge Systems, ICIKS 2023, Proceedings",

address = "United States",

}

Manai, M & Ben Yahia, S 2024, Efficient Topic Detection Using an Adaptive Neural Network Architecture. i I Saad, C Rosenthal-Sabroux, F Gargouri, S Chakhar, N Williams & E Haig (red), Advances in Information Systems, Artificial Intelligence and Knowledge Management - 6th International Conference on Information and Knowledge Systems, ICIKS 2023, Proceedings. Springer Science+Business Media, Lecture Notes in Business Information Processing, bind 486, s. 145-157, 6th International Conference on Information and Knowledge Systems, ICIKS 2023, Portsmouth, Storbritannien, 22/06/2023. https://doi.org/10.1007/978-3-031-51664-1_10

Efficient Topic Detection Using an Adaptive Neural Network Architecture. / Manai, Meriem; Ben Yahia, Sadok.
Advances in Information Systems, Artificial Intelligence and Knowledge Management - 6th International Conference on Information and Knowledge Systems, ICIKS 2023, Proceedings. red. / Inès Saad; Camille Rosenthal-Sabroux; Faiez Gargouri; Salem Chakhar; Nigel Williams; Ella Haig. Springer Science+Business Media, 2024. s. 145-157 (Lecture Notes in Business Information Processing, Bind 486).

Publikation: Kapitel i bog/rapport/konference-proceeding › Konferencebidrag i proceedings › Forskning › peer review

TY - GEN

T1 - Efficient Topic Detection Using an Adaptive Neural Network Architecture

AU - Manai, Meriem

AU - Ben Yahia, Sadok

N1 - .

PY - 2024

Y1 - 2024

N2 - Topic detection is the process of identifying the underlying themes or topics present in a set of documents. It has become more critical due to the increase of information electronically available and the necessity to process and filter it. In this respect, we introduce a new approach to detecting topics called ClusART. Thus, we created a three-phase approach: a first phase during which lexical preprocessing was conducted. The second phase pays heed to the construction and the generation of vectors representing the documents carried out. In the topic detection phase, the FuzzyART algorithm is used for the training phase, and a classifier based on ParagraphVector is used for the test phase. The comparative study of our approach on the 20 Newsgroups dataset showed that our method could detect almost relevant topics.

AB - Topic detection is the process of identifying the underlying themes or topics present in a set of documents. It has become more critical due to the increase of information electronically available and the necessity to process and filter it. In this respect, we introduce a new approach to detecting topics called ClusART. Thus, we created a three-phase approach: a first phase during which lexical preprocessing was conducted. The second phase pays heed to the construction and the generation of vectors representing the documents carried out. In the topic detection phase, the FuzzyART algorithm is used for the training phase, and a classifier based on ParagraphVector is used for the test phase. The comparative study of our approach on the 20 Newsgroups dataset showed that our method could detect almost relevant topics.

KW - FuzzyART

KW - PragraphVector classifier

KW - Text clustering

KW - Topic detection

U2 - 10.1007/978-3-031-51664-1_10

DO - 10.1007/978-3-031-51664-1_10

M3 - Article in proceedings

AN - SCOPUS:85184279177

SN - 9783031516634

T3 - Lecture Notes in Business Information Processing

SP - 145

EP - 157

BT - Advances in Information Systems, Artificial Intelligence and Knowledge Management - 6th International Conference on Information and Knowledge Systems, ICIKS 2023, Proceedings

A2 - Saad, Inès

A2 - Rosenthal-Sabroux, Camille

A2 - Gargouri, Faiez

A2 - Chakhar, Salem

A2 - Williams, Nigel

A2 - Haig, Ella

PB - Springer Science+Business Media

T2 - 6th International Conference on Information and Knowledge Systems, ICIKS 2023

Y2 - 22 June 2023 through 23 June 2023

ER -

Manai M, Ben Yahia S. Efficient Topic Detection Using an Adaptive Neural Network Architecture. I Saad I, Rosenthal-Sabroux C, Gargouri F, Chakhar S, Williams N, Haig E, red., Advances in Information Systems, Artificial Intelligence and Knowledge Management - 6th International Conference on Information and Knowledge Systems, ICIKS 2023, Proceedings. Springer Science+Business Media. 2024. s. 145-157. (Lecture Notes in Business Information Processing, Bind 486). doi: 10.1007/978-3-031-51664-1_10