Pretražite po imenu i prezimenu autora, mentora, urednika, prevoditelja

Napredna pretraga

Pregled bibliografske jedinice broj: 1143675

A Topic Coverage Approach to Evaluation of Topic Models


Korenčić, Damir; Ristov, Strahil; Repar, Jelena; Šnajder, Jan
A Topic Coverage Approach to Evaluation of Topic Models // IEEE access, 9 (2021), 123280-123312 doi:10.1109/access.2021.3109425 (međunarodna recenzija, članak, znanstveni)


CROSBI ID: 1143675 Za ispravke kontaktirajte CROSBI podršku putem web obrasca

Naslov
A Topic Coverage Approach to Evaluation of Topic Models

Autori
Korenčić, Damir ; Ristov, Strahil ; Repar, Jelena ; Šnajder, Jan

Izvornik
IEEE access (2169-3536) 9 (2021); 123280-123312

Vrsta, podvrsta i kategorija rada
Radovi u časopisima, članak, znanstveni

Ključne riječi
Topic coverage ; Topic coherence ; Topic discovery ; Topic models ; Topic model evaluation ; Topic model stability

Sažetak
Topic models are widely used unsupervised models capable of learning topics – weightedlists of words and documents – from large collections of text documents. When topic models are used fordiscovery of topics in text collections, a question that arises naturally is how well the model-induced topicscorrespond to topics of interest to the analyst. In this paper we revisit and extend a so far neglected approachto topic model evaluation based on measuring topic coverage – computationally matching model topics witha set of reference topics that models are expected to uncover. The approach is well suited for analyzingmodels’ performance in topic discovery and for large-scale analysis of both topic models and measures ofmodel quality. We propose new measures of coverage and evaluate, in a series of experiments, different typesof topic models on two distinct text domains for which interest for topic discovery exists. The experimentsinclude evaluation of model quality, analysis of coverage of distinct topic categories, and the analysis of therelationship between coverage and other methods of topic model evaluation. The paper contributes a newsupervised measure of coverage, and the first unsupervised measure of coverage. The supervised measureachieves topic matching accuracy close to human agreement. The unsupervised measure correlates highlywith the supervised one (Spearman’s ρ≥ 0.95). Other contributions include insights into both topic modelsand different methods of model evaluation, and the datasets and code for facilitating future research on topiccoverage.

Izvorni jezik
Engleski

Znanstvena područja
Računarstvo



POVEZANOST RADA


Projekti:
EK-KF-KK.01.1.1.01.0009-1 - Napredne metode i tehnologije u znanosti o podatcima i kooperativnim sustavima - IJ za znanost o podatcima (Lončarić, Sven, EK - KK.01.1.1.01) ( POIROT)

Ustanove:
Fakultet elektrotehnike i računarstva, Zagreb,
Institut "Ruđer Bošković", Zagreb

Profili:

Avatar Url Jelena Repar (autor)

Avatar Url Damir Korenčić (autor)

Avatar Url Jan Šnajder (autor)

Avatar Url Strahil Ristov (autor)

Poveznice na cjeloviti tekst rada:

doi ieeexplore.ieee.org

Citiraj ovu publikaciju:

Korenčić, Damir; Ristov, Strahil; Repar, Jelena; Šnajder, Jan
A Topic Coverage Approach to Evaluation of Topic Models // IEEE access, 9 (2021), 123280-123312 doi:10.1109/access.2021.3109425 (međunarodna recenzija, članak, znanstveni)
Korenčić, D., Ristov, S., Repar, J. & Šnajder, J. (2021) A Topic Coverage Approach to Evaluation of Topic Models. IEEE access, 9, 123280-123312 doi:10.1109/access.2021.3109425.
@article{article, year = {2021}, pages = {123280-123312}, DOI = {10.1109/access.2021.3109425}, keywords = {Topic coverage, Topic coherence, Topic discovery, Topic models, Topic model evaluation, Topic model stability}, journal = {IEEE access}, doi = {10.1109/access.2021.3109425}, volume = {9}, issn = {2169-3536}, title = {A Topic Coverage Approach to Evaluation of Topic Models}, keyword = {Topic coverage, Topic coherence, Topic discovery, Topic models, Topic model evaluation, Topic model stability} }
@article{article, year = {2021}, pages = {123280-123312}, DOI = {10.1109/access.2021.3109425}, keywords = {Topic coverage, Topic coherence, Topic discovery, Topic models, Topic model evaluation, Topic model stability}, journal = {IEEE access}, doi = {10.1109/access.2021.3109425}, volume = {9}, issn = {2169-3536}, title = {A Topic Coverage Approach to Evaluation of Topic Models}, keyword = {Topic coverage, Topic coherence, Topic discovery, Topic models, Topic model evaluation, Topic model stability} }

Časopis indeksira:


  • Current Contents Connect (CCC)
  • Web of Science Core Collection (WoSCC)
    • Science Citation Index Expanded (SCI-EXP)
    • SCI-EXP, SSCI i/ili A&HCI
  • Scopus


Citati:





    Contrast
    Increase Font
    Decrease Font
    Dyslexic Font