Browsing by Author "Pung, Andreas"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    Konvolutsioonilisel neurovõrgul põhineva teksti klassifitseerimismudeli interpreteerimine kliinilisel andmestikul
    (2019) Pung, Andreas; Kairit Sirts
    Bakalaureusetöös interpreteeritakse konvolutsioonilist teksti klassifitseerimise neurovõrku: miks just niimoodi neurovõrk klassifikatsiooniotsuseid teeb. Analüüsi teostati kliinilisel DementiaBanki andmestikul, kus on Alzheimeri tõvega inimeste kirjeldused Bostoni küpsisevarguse fotost. Binaarse klassifikatsiooni ülesanne seisnes etteantud teksti põhjal klassi tuvastamises: kas isikul on Alzheimeri tõbi või mitte. Programmeeriti valmis Jacovi et al. artiklis (2018) kirjeldatud interpretatsioonimeetodid. Samuti interpreteeritakse konkreetseid tekste. Interpretatsioonimeetoditest andsid hea tulemuse informatiivsete ja ebainformatiivsete n-grammide leidmine, sõnede aktivatsioonivektorite leidmine ning nende klasterdamine. Vastandlike n-grammide analüüs andis halvema tulemuse andmestiku eripärade tõttu.
  • Thumbnail Image
    Item
    Weakly-Supervised Text Classification for Estonian Sentiment Analysis
    (Tartu Ülikool, 2022) Pung, Andreas; Sirts, Kairit, juhendaja; Tartu Ülikool. Loodus- ja täppisteaduste valdkond; Tartu Ülikool. Arvutiteaduse instituut
    Text Classification is one of the most fundamental tasks in Natural Language Processing. Hand-labelling texts is costly and might need specialised domain knowledge – this is where unsupervised and weakly-supervised approaches could be useful. In this Master’s Thesis, the weakly-supervised text classification paradigm is used to classify the sentiment of Estonian texts. In this paradigm, the weak labels are created using labelling functions (Ratner et al., 2016). The aim of this thesis is to assess the applicability of weakly-supervised models trained with around 40× larger dataset in contrast to hand-labelling a smaller amount of texts to train a fully-supervised classifier. The compared models are fully and weaklysupervised BERT (Devlin et al., 2019); weakly-supervised COSINE (Yu et al., 2021) and WeaSEL (Cachay et al., 2021). Human evaluation is performed on texts where the models disagreed the most. As a result, we find that the fully-supervised models have the best performance. The best-performing weakly-supervised model trained on the larger dataset had an average classification accuracy of 7.29% worse (7.05% worse weighted F1-score) than the fully-supervised BERT model. The lower performance of weakly-supervised models might be caused by the low quality of labelling functions – developing them further might lead to better results.