Andraž Pelicon


2024

pdf
Denoising Labeled Data for Comment Moderation Using Active Learning
Andraž Pelicon | Mladen Karan | Ravi Shekhar | Matthew Purver | Senja Pollak
Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)

Noisily labeled textual data is ample on internet platforms that allow user-created content. Training models, such as offensive language detection models for comment moderation, on such data may prove difficult as the noise in the labels prevents the model to converge. In this work, we propose to use active learning methods for the purposes of denoising training data for model training. The goal is to sample examples the most informative examples with noisy labels with active learning and send them to the oracle for reannotation thus reducing the overall cost of reannotation. In this setting we tested three existing active learning methods, namely DBAL, Variance of Gradients (VoG) and BADGE. The proposed approach to data denoising is tested on the problem of offensive language detection. We observe that active learning can be effectively used for the purposes of data denoising, however care should be taken when choosing the algorithm for this purpose.

2022

pdf
Embeddings models for Buddhist Sanskrit
Ligeia Lugli | Matej Martinc | Andraž Pelicon | Senja Pollak
Proceedings of the Thirteenth Language Resources and Evaluation Conference

The paper presents novel resources and experiments for Buddhist Sanskrit, broadly defined here including all the varieties of Sanskrit in which Buddhist texts have been transmitted. We release a novel corpus of Buddhist texts, a novel corpus of general Sanskrit and word similarity and word analogy datasets for intrinsic evaluation of Buddhist Sanskrit embeddings models. We compare the performance of word2vec and fastText static embeddings models, with default and optimized parameter settings, as well as contextual models BERT and GPT-2, with different training regimes (including a transfer learning approach using the general Sanskrit corpus) and different embeddings construction regimes (given the encoder layers). The results show that for semantic similarity the fastText embeddings yield the best results, while for word analogy tasks BERT embeddings work the best. We also show that for contextual models the optimal layer combination for embedding construction is task dependant, and that pretraining the contextual embeddings models on a reference corpus of general Sanskrit is beneficial, which is a promising finding for future development of embeddings for less-resourced languages and domains.

pdf
EMBEDDIA project: Cross-Lingual Embeddings for Less- Represented Languages in European News Media
Senja Pollak | Andraž Pelicon
Proceedings of the 23rd Annual Conference of the European Association for Machine Translation

EMBEDDIA project developed a range of resources and methods for less-resourced EU languages, focusing on applications for media industry, including keyword extraction, comment moderation and article generation.

pdf
Sentiment Classification by Incorporating Background Knowledge from Financial Ontologies
Timen Stepišnik-Perdih | Andraž Pelicon | Blaž Škrlj | Martin Žnidaršič | Igor Lončarski | Senja Pollak
Proceedings of the 4th Financial Narrative Processing Workshop @LREC2022

Ontologies are increasingly used for machine reasoning over the last few years. They can provide explanations of concepts or be used for concept classification if there exists a mapping from the desired labels to the relevant ontology. This paper presents a practical use of an ontology for the purpose of data set generalization in an oversampling setting, with the aim of improving classification models. We demonstrate our solution on a novel financial sentiment data set using the Financial Industry Business Ontology (FIBO). The results show that generalization-based data enrichment benefits simpler models in a general setting and more complex models such as BERT in low-data setting.

2021

pdf
Zero-shot Cross-lingual Content Filtering: Offensive Language and Hate Speech Detection
Andraž Pelicon | Ravi Shekhar | Matej Martinc | Blaž Škrlj | Matthew Purver | Senja Pollak
Proceedings of the EACL Hackashop on News Media Content Analysis and Automated Report Generation

We present a system for zero-shot cross-lingual offensive language and hate speech classification. The system was trained on English datasets and tested on a task of detecting hate speech and offensive social media content in a number of languages without any additional training. Experiments show an impressive ability of both models to generalize from English to other languages. There is however an expected gap in performance between the tested cross-lingual models and the monolingual models. The best performing model (offensive content classifier) is available online as a REST API.

pdf
EMBEDDIA Tools, Datasets and Challenges: Resources and Hackathon Contributions
Senja Pollak | Marko Robnik-Šikonja | Matthew Purver | Michele Boggia | Ravi Shekhar | Marko Pranjić | Salla Salmela | Ivar Krustok | Tarmo Paju | Carl-Gustav Linden | Leo Leppänen | Elaine Zosa | Matej Ulčar | Linda Freienthal | Silver Traat | Luis Adrián Cabrera-Diego | Matej Martinc | Nada Lavrač | Blaž Škrlj | Martin Žnidaršič | Andraž Pelicon | Boshko Koloski | Vid Podpečan | Janez Kranjc | Shane Sheehan | Emanuela Boros | Jose G. Moreno | Antoine Doucet | Hannu Toivonen
Proceedings of the EACL Hackashop on News Media Content Analysis and Automated Report Generation

This paper presents tools and data sources collected and released by the EMBEDDIA project, supported by the European Union’s Horizon 2020 research and innovation program. The collected resources were offered to participants of a hackathon organized as part of the EACL Hackashop on News Media Content Analysis and Automated Report Generation in February 2021. The hackathon had six participating teams who addressed different challenges, either from the list of proposed challenges or their own news-industry-related tasks. This paper goes beyond the scope of the hackathon, as it brings together in a coherent and compact form most of the resources developed, collected and released by the EMBEDDIA project. Moreover, it constitutes a handy source for news media industry and researchers in the fields of Natural Language Processing and Social Science.

pdf
EMBEDDIA hackathon report: Automatic sentiment and viewpoint analysis of Slovenian news corpus on the topic of LGBTIQ+
Matej Martinc | Nina Perger | Andraž Pelicon | Matej Ulčar | Andreja Vezovnik | Senja Pollak
Proceedings of the EACL Hackashop on News Media Content Analysis and Automated Report Generation

We conduct automatic sentiment and viewpoint analysis of the newly created Slovenian news corpus containing articles related to the topic of LGBTIQ+ by employing the state-of-the-art news sentiment classifier and a system for semantic change detection. The focus is on the differences in reporting between quality news media with long tradition and news media with financial and political connections to SDS, a Slovene right-wing political party. The results suggest that political affiliation of the media can affect the sentiment distribution of articles and the framing of specific LGBTIQ+ specific topics, such as same-sex marriage.

2019

pdf
Embeddia at SemEval-2019 Task 6: Detecting Hate with Neural Network and Transfer Learning Approaches
Andraž Pelicon | Matej Martinc | Petra Kralj Novak
Proceedings of the 13th International Workshop on Semantic Evaluation

SemEval 2019 Task 6 was OffensEval: Identifying and Categorizing Offensive Language in Social Media. The task was further divided into three sub-tasks: offensive language identification, automatic categorization of offense types, and offense target identification. In this paper, we present the approaches used by the Embeddia team, who qualified as fourth, eighteenth and fifth on the tree sub-tasks. A different model was trained for each sub-task. For the first sub-task, we used a BERT model fine-tuned on the OLID dataset, while for the second and third tasks we developed a custom neural network architecture which combines bag-of-words features and automatically generated sequence-based features. Our results show that combining automatically and manually crafted features fed into a neural architecture outperform transfer learning approach on more unbalanced datasets.