Knowledge graph informed fake news classification via heterogeneous representation ensembles B Koloski, TS Perdih, M Robnik-Šikonja, S Pollak, B Škrlj Neurocomputing 496, 208-226, 2022 | 68 | 2022 |
BERT meets shapley: Extending SHAP explanations to transformer-based classifiers E Kokalj, B Škrlj, N Lavrač, S Pollak, M Robnik-Šikonja Proceedings of the EACL hackashop on news media content analysis and …, 2021 | 60 | 2021 |
tax2vec: Constructing interpretable features from taxonomies for short text classification B Škrlj, M Martinc, J Kralj, N Lavrač, S Pollak Computer Speech & Language 65, 101104, 2021 | 58 | 2021 |
Zero-shot learning for cross-lingual news sentiment classification A Pelicon, M Pranjić, D Miljković, B Škrlj, S Pollak Applied Sciences 10 (17), 5993, 2020 | 55 | 2020 |
TNT-KID: Transformer-based neural tagger for keyword identification M Martinc, B Škrlj, S Pollak Natural Language Engineering 28 (4), 409-448, 2022 | 49 | 2022 |
Feature importance estimation with self-attention networks B Škrlj, S Džeroski, N Lavrač, M Petkovič ECAI 2020 325, 1491--1498, 2020 | 47 | 2020 |
To ban or not to ban: Bayesian attention networks for reliable hate speech detection K Miok, B Škrlj, D Zaharie, M Robnik-Šikonja Cognitive Computation 14 (1), 353-371, 2022 | 43 | 2022 |
Propositionalization and embeddings: two sides of the same coin N Lavrač, B Škrlj, M Robnik-Šikonja Machine Learning 109, 1465-1507, 2020 | 37 | 2020 |
RaKUn: Rank-based Keyword Extraction via Unsupervised Learning and Meta Vertex Aggregation B Škrlj, A Repar, S Pollak Statistical Language and Speech Processing: 7th International Conference …, 2019 | 37 | 2019 |
Prioritisation of Compounds for 3CLpro Inhibitor Development on SARS-CoV-2 Variants M Jukič, B Škrlj, G Tomšič, S Pleško, Č Podlipnik, U Bren Molecules 26 (10), 3003, 2021 | 28 | 2021 |
Towards robust text classification with semantics-aware recurrent neural architecture B Škrlj, J Kralj, N Lavrač, S Pollak Machine Learning and Knowledge Extraction 1 (2), 34, 2019 | 28 | 2019 |
Investigating cross-lingual training for offensive language detection A Pelicon, R Shekhar, B Škrlj, M Purver, S Pollak PeerJ Computer Science 7, e559, 2021 | 27 | 2021 |
Identification of COVID-19 related fake news via neural stacking B Koloski, T Stepišnik-Perdih, S Pollak, B Škrlj Combating Online Hostile Posts in Regional Languages during Emergency …, 2021 | 25 | 2021 |
Prediction uncertainty estimation for hate speech classification K Miok, D Nguyen-Doan, B Škrlj, D Zaharie, M Robnik-Šikonja Statistical Language and Speech Processing: 7th International Conference …, 2019 | 25 | 2019 |
A comprehensive comparison of molecular feature representations for use in predictive modeling T Stepišnik, B Škrlj, J Wicker, D Kocev Computers in Biology and Medicine 130, 104197, 2021 | 23 | 2021 |
ProBiS-Dock Database: a web server and interactive web repository of small ligand–protein binding sites for drug design J Konc, S Lesnik, B Skrlj, D Janezic Journal of chemical information and modeling 61 (8), 4097-4107, 2021 | 22 | 2021 |
The role of the Listeria monocytogenes surfactome in biofilm formation N Janež, B Škrlj, M Sterniša, A Klančnik, J Sabotič Microbial biotechnology 14 (4), 1269-1281, 2021 | 21 | 2021 |
Py3plex toolkit for visualization and analysis of multilayer networks B Škrlj, J Kralj, N Lavrač Applied Network Science 4 (1), 1-24, 2019 | 20 | 2019 |
autoBOT: evolving neuro-symbolic representations for explainable low resource text classification B Škrlj, M Martinc, N Lavrač, S Pollak Machine Learning 110 (5), 989-1028, 2021 | 19 | 2021 |
Extending neural keyword extraction with TF-IDF tagset matching B Koloski, S Pollak, B Škrlj, M Martinc arXiv preprint arXiv:2102.00472, 2021 | 18 | 2021 |