ŞÖŞaziye Betül Özateş

Şaziye Betül Özateş

*This is an algorithmically generated author record based on publicly available information.

Şaziye Betül Özateş has a total of 15 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Learning from Rare Clicks: A Comparative Study of Imbalance-Handling Techniques For Deep CTR Prediction

    by Tugay Balatlı, Şaziye Betül Özateş, Ercan Atam

    Click-through Rate (CTR) prediction is an essential but difficult task in online advertising and recommendation systems. The imbalanced nature of click data, where positive click instances are much fewer in number than non-click instances, is one of the major issues in this task. This work presents an extensive comparison of rare event handling methods combined with three state-of-the-art deep learning models for CTR prediction tasks, namely, Deep and Cross Network (DCN), Deep Factorization Machine (DeepFM), and Deep Neural Network (DNN). The rare event handling methods being considered are data-level methods such as Synthetic Minority Over-sampling Technique (SMOTE) and NearMiss undersampling, and algorithm-level methods such as class weight, PolyLoss, and Dice loss functions. The experiments are conducted on two large real-world datasets: the Avazu CTR prediction dataset, containing 1.4 million training samples and a CTR of 16.16 percent; and the Trendyol JFY Ads dataset, a 5 million training sample dataset provided by the company Trendyol, an e-commerce company, where the CTR is around 5 percent. The experimental results, validated through multi-seed Wilcoxon signed-rank tests ($n = 7$, $p < 0.05$), show that PolyLoss achieves the best mean AUC on Avazu (0.7675 with DNN, +0.44\% over baseline) and on Trendyol (0.6710 with DCN, +0.56\% over baseline). These results demonstrate that algorithm-level methods show greater consistency in performance improvements across diverse datasets and models, compared to data-level approaches.

  • A Deep Learning Approach for Gain-Scheduled PID Control Design via Advanced Supervisory Controllers

    by Ercan Atam, Atakan Zeybek, Şaziye Betül Özateş, Nehir Güzelkaya

    Proportional-integral-derivative (PID) control is by far the most widely used control type in industry. However, for some systems even the optimally tuned PID control system can be unsatisfactory, especially when the underlying system is nonlinear, includes parameter-varying components and time delays or when system dynamics change with time. Due to the long-standing development and readiness of the infrastructure of PID controllers in the industry, along with cost factors, it is essential to improve the system performance without replacing the established structure. Gain-scheduled PID control is a strong solution for such situations. In this paper, we present a new approach for the gain-scheduled PID control design using deep learning to learn from advanced controllers such as linear parameter-varying controller and reinforcement learning agent. For each gain of PID, a deep artificial neural network is developed as a function of the scheduling signals. To validate the proposed approach, we tested it on two challenging case studies and compared its performance against optimally tuned conventional PID controllers with fixed parameters, as well as advanced state-of-the-art controllers that the gain-scheduled PID controllers aim to emulate. The proposed gain-scheduled method provided up to 71.6% performance improvement relative to the optimally tuned PID controller with fixed gains.


  • Open Access

    A Hybrid Deep Dependency Parsing Approach Enhanced With Rules and Morphology: A Case Study for Turkish

    by Saziye Betul Ozates, Arzucan Ozgur, Tunga Gungor, Balkiz Ozturk Basaran

    Fully data-driven, deep learning-based models are usually designed as language-independent and have been shown to be successful for many natural language processing tasks. However, when the studied language is not high-resource and the amount of training data is insufficient, these models can benefit from the integration of natural language grammar-based information. We propose two approaches to dependency parsing especially for languages with restricted amount of training data. 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. In the rule-based approach, the parsing decisions made by the rules are encoded and concatenated with the vector representations of the input words as additional information to the deep network. The morphology-based approach proposes different methods to include the morphological structure of words into the parser network. Experiments are conducted on three different Turkish treebanks and the results suggest that integration of explicit knowledge about the target language to a neural parser through a rule-based parsing system and morphological analysis leads to more accurate annotations and hence, increases the parsing performance in terms of attachment scores. The proposed methods are developed for Turkish, but can be adapted to other languages as well.


  • Resources for Turkish dependency parsing: introducing the BOUN Treebank and the BoAT annotation tool

    by Utku Türk, Furkan Atmaca, Şaziye Betül Özateş, Gözde Berk, Seyyit Talha Bedir, Abdullatif Köksal, Balkız Öztürk Başaran, Tunga Güngör, Arzucan Özgür


  • Open Access

    A Hybrid Approach to Dependency Parsing: Combining Rules and Morphology with Deep Learning

    by Şaziye Betül Özateş, Arzucan Özgür, Tunga Güngör, Balkız Öztürk

    Fully data-driven, deep learning-based models are usually designed as language-independent and have been shown to be successful for many natural language processing tasks. However, when the studied language is low-resourced and the amount of training data is insufficient, these models can benefit from the integration of natural language grammar-based information. We propose two approaches to dependency parsing especially for languages with restricted amount of training data. 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. In the rule-based approach, the parsing decisions made by the rules are encoded and concatenated with the vector representations of the input words as additional information to the deep network. The morphology-based approach proposes different methods to include the morphological structure of words into the parser network. Experiments are conducted on the IMST-UD Treebank and the results suggest that integration of explicit knowledge about the target language to a neural parser through a rule-based parsing system and morphological analysis leads to more accurate annotations and hence, increases the parsing performance in terms of attachment scores. The proposed methods are developed for Turkish, but can be adapted to other languages as well.

  • Open Access

    Turkish Treebanking: Unifying and Constructing Efforts

    by Utku Türk, Furkan Atmaca, Şaziye Betül Özateş, Abdullatif Köksal, Balkiz Ozturk Basaran, Tunga Gungor, Arzucan Özgür