no code implementations • LATERAISSE (LREC) 2022 • Zehra Melce Hüsünbeyi, Didar Akar, Arzucan Özgür
We studied the compatibility of our model with the hate speech detection problem by comparing it with traditional machine learning models, as well as a Convolution Neural Network (CNN) based model, a Convolutional Neural Network-Gated Recurrent Unit (CNN-GRU) based model which reached significant performance results for hate speech detection.
1 code implementation • EMNLP (NLP-COVID19) 2020 • Abdullatif Köksal, Hilal Dönmez, Rıza Özçelik, Elif Ozkirimli, Arzucan Özgür
Coronavirus Disease of 2019 (COVID-19) created dire consequences globally and triggered an intense scientific effort from different domains.
1 code implementation • Findings (NAACL) 2022 • Şaziye Özateş, Arzucan Özgür, Tunga Gungor, Özlem Çetinoğlu
Code-switching dependency parsing stands as a challenging task due to both the scarcity of necessary resources and the structural difficulties embedded in code-switched languages.
no code implementations • EMNLP (LAW, DMR) 2021 • Talha Bedir, Karahan Şahin, Onur Gungor, Suzan Uskudarli, Arzucan Özgür, Tunga Güngör, Balkiz Ozturk Basaran
This paper presents these issues and our proposals to more accurately represent morphosyntactic information for Turkish while adhering to guidelines of UD.
no code implementations • 30 Mar 2023 • Hasin Rehana, Nur Bengisu Çam, Mert Basmaci, Jie Zheng, Christianah Jemiyo, Yongqun He, Arzucan Özgür, Junguk Hur
It achieved a precision of 88. 37%, a recall of 85. 14%, and an F1-score of 86. 49% on the LLL dataset.
1 code implementation • 5 Jan 2023 • Gonul Ayci, Arzucan Özgür, Murat Şensoy, Pinar Yolum
The generated explanations can be used by users to understand the recommendations of the privacy assistant.
no code implementations • 26 Oct 2022 • Asu Büşra Temizer, Gökçe Uludoğan, Rıza Özçelik, Taha Koulani, Elif Ozkirimli, Kutlu O. Ulgen, Nilgün Karalı, Arzucan Özgür
To this end, we build a language-inspired pipeline that treats high affinity ligands of protein targets as documents and selects key chemical words making up those ligands based on tf-idf weighting.
1 code implementation • 2 Sep 2022 • Gökçe Uludoğan, Elif Ozkirimli, Kutlu O. Ulgen, Nilgün Karalı, Arzucan Özgür
On the other hand, large amounts of unlabeled protein sequences and chemical compounds are available and have been used to train language models that learn useful representations.
no code implementations • 24 Jul 2022 • Büşra Marşan, Salih Furkan Akkurt, Muhammet Şen, Merve Gürbüz, Onur Güngör, Şaziye Betül Özateş, Suzan Üsküdarlı, Arzucan Özgür, Tunga Güngör, Balkız Öztürk
In this study, we aim to offer linguistically motivated solutions to resolve the issues of the lack of representation of null morphemes, highly productive derivational processes, and syncretic morphemes of Turkish in the BOUN Treebank without diverging from the Universal Dependencies framework.
no code implementations • 13 May 2022 • Gonul Ayci, Murat Sensoy, Arzucan Özgür, Pinar Yolum
By factoring in the user's own understanding of privacy, such as risk factors or own labels, the personal assistant can personalize its recommendations per user.
no code implementations • ACL 2022 • M. Melih Mutlu, Arzucan Özgür
For low-resource languages such as Turkish, there is a lack of such annotated data.
no code implementations • 23 Sep 2021 • Arda Çelebi, Arzucan Özgür
At the named entity disambiguation phase, first the cluster-based types of a given mention are predicted and then, these types are used as features in a ranking model to select the best entity among the candidates.
2 code implementations • EMNLP 2021 • Yi Huang, Buse Giledereli, Abdullatif Köksal, Arzucan Özgür, Elif Ozkirimli
Here, we introduce the application of balancing loss functions for multi-label text classification.
Ranked #1 on Multi-Label Text Classification on Reuters-21578
3 code implementations • 4 Jul 2021 • Rıza Özçelik, Alperen Bağ, Berk Atıl, Melih Barsbey, Arzucan Özgür, Elif Özkırımlı
Here, we present DebiasedDTA, a novel drug-target affinity (DTA) prediction model training framework that addresses dataset biases to improve the generalizability of affinity prediction models.
1 code implementation • 19 Oct 2020 • Abdullatif Köksal, Arzucan Özgür
Relation classification is one of the key topics in information extraction, which can be used to construct knowledge bases or to provide useful information for question answering.
no code implementations • 5 Sep 2020 • Abdullatif Köksal, Hilal Dönmez, Rıza Özçelik, Elif Ozkirimli, Arzucan Özgür
Coronavirus Disease of 2019 (COVID-19) created dire consequences globally and triggered an intense scientific effort from different domains.
no code implementations • 24 Feb 2020 • Şaziye Betül Özateş, Arzucan Özgür, Tunga Güngör, Balkız Öztürk
Our first approach combines a state-of-the-art deep learning-based parser with a rule-based approach and the second one incorporates morphological information into the parser.
1 code implementation • 24 Feb 2020 • Utku Türk, Furkan Atmaca, Şaziye Betül Özateş, Gözde Berk, Seyyit Talha Bedir, Abdullatif Köksal, Balkız Öztürk Başaran, Tunga Güngör, Arzucan Özgür
In addition, we report the parsing results of a state-of-the-art dependency parser obtained over the BOUN Treebank as well as two other treebanks in Turkish.
Cultural Vocal Bursts Intensity Prediction Dependency Parsing
no code implementations • 10 Feb 2020 • Hakime Öztürk, Arzucan Özgür, Philippe Schwaller, Teodoro Laino, Elif Ozkirimli
Text-based representations of chemicals and proteins can be thought of as unstructured languages codified by humans to describe domain-specific knowledge.
no code implementations • 4 Feb 2019 • Hakime Öztürk, Elif Ozkirimli, Arzucan Özgür
In addition, the results showed that, given the protein sequence and ligand SMILES, the inclusion of protein domain and motif information as well as ligand maximum common substructure words do not provide additional useful information for the deep learning model.
1 code implementation • 2 Nov 2018 • Rıza Özçelik, Hakime Öztürk, Arzucan Özgür, Elif Ozkirimli
Our aim is to process the patterns in SMILES as a language to predict protein-ligand affinity, even when we cannot infer the function from the sequence.
no code implementations • LREC 2016 • Eda Okur, Hakan Demir, Arzucan Özgür
We made use of these obtained word embeddings, together with language independent features that are engineered to work better on informal text types, for generating a Turkish NER system on microblog texts.
no code implementations • 30 Jan 2018 • Hakime Öztürk, Elif Ozkirimli, Arzucan Özgür
We showed that ligand-based protein representation, which uses only SMILES strings of the ligands that proteins bind to, performs as well as protein-sequence based representation methods in protein clustering.
4 code implementations • 30 Jan 2018 • Hakime Öztürk, Elif Ozkirimli, Arzucan Özgür
The results show that the proposed deep learning based model that uses the 1D representations of targets and drugs is an effective approach for drug target binding affinity prediction.
Ranked #1 on Drug Discovery on BindingDB IC50
no code implementations • Bioinformatics 2017 • Gizem Sogancioglu, Hakime Öztürk, Arzucan Özgür
A benchmark data set consisting of 100 sentence pairs from the biomedical literature is manually annotated by five human experts and used for evaluating the proposed methods.
Ranked #8 on Sentence Embeddings For Biomedical Texts on BIOSSES (using extra training data)