Cristin-resultat-ID: 2081439
Sist endret: 8. desember 2022, 10:38
NVI-rapporteringsår: 2022
Resultat
Vitenskapelig Kapittel/Artikkel/Konferanseartikkel
2022

ConvTextTM: An Explainable Convolutional Tsetlin Machine Framework for Text Classification

Bidragsytere:
  • Bimal Bhattarai
  • Ole-Christoffer Granmo og
  • Jiao Lei

Bok

Proceedings of the Thirteenth Language Resources and Evaluation Conference
ISBN:
  • 979-10-95546-72-6

Utgiver

European Language Resources Association
NVI-nivå 1

Om resultatet

Vitenskapelig Kapittel/Artikkel/Konferanseartikkel
Publiseringsår: 2022
Sider: 3761 - 3770
ISBN:
  • 979-10-95546-72-6

Klassifisering

Fagfelt (NPI)

Fagfelt: IKT
- Fagområde: Realfag og teknologi

Beskrivelse Beskrivelse

Tittel

ConvTextTM: An Explainable Convolutional Tsetlin Machine Framework for Text Classification

Sammendrag

Recent advancements in natural language processing (NLP) have reshaped the industry, with powerful language models such as GPT-3 achieving superhuman performance on various tasks. However, the increasing complexity of such models turns them into “black boxes”, creating uncertainty about their internal operation and decision-making. Tsetlin Machine (TM) employs human-interpretable conjunctive clauses in propositional logic to solve complex pattern recognition problems and has demonstrated competitive performance in various NLP tasks. In this paper, we propose ConvTextTM, a novel convolutional TM architecture for text classification. While legacy TM solutions treat the whole text as a corpus-specific set-of-words (SOW), ConvTextTM breaks down the text into a sequence of text fragments. The convolution over the text fragments opens up for local position-aware analysis. Further, ConvTextTM eliminates the dependency on a corpus-specific vocabulary. Instead, it employs a generic SOW formed by the tokenization scheme of the Bidirectional Encoder Representations from Transformers (BERT) (Devlin et al., 2019a). The convolution binds together the tokens, allowing ConvTextTM to address the out-of-vocabulary problem as well as spelling errors. We investigate the local explainability of our proposed method using clause-based features. Extensive experiments are conducted on seven datasets, to demonstrate that the accuracy of ConvTextTM is either superior or comparable to state-of-the-art baselines.

Bidragsytere

Bimal Bhattarai

  • Tilknyttet:
    Forfatter
    ved Institutt for informasjons- og kommunikasjonsteknologi ved Universitetet i Agder

Ole-Christoffer Granmo

  • Tilknyttet:
    Forfatter
    ved Institutt for informasjons- og kommunikasjonsteknologi ved Universitetet i Agder

Lei Jiao

Bidragsyterens navn vises på dette resultatet som Jiao Lei
  • Tilknyttet:
    Forfatter
    ved Institutt for informasjons- og kommunikasjonsteknologi ved Universitetet i Agder
1 - 3 av 3

Resultatet er en del av Resultatet er en del av

Proceedings of the Thirteenth Language Resources and Evaluation Conference.

Calzolari, Nicoletta; Béchet, Frédéric; Blache, Philippe; Choukri, Khalid; Cieri, Christopher; Declerck, Thierry; Goggi, Sara; Isahara, Hitoshi; Maegaard, Bente; Mariani, Joseph mfl.. 2022, European Language Resources Association. Vitenskapelig antologi/Konferanseserie
1 - 1 av 1