Yazar "Balci, Emirhan" seçeneğine göre listele

Listeleniyor 1 - 3 / 3
  • [ X ]
    Öğe
    Automated Depression Detection from Tweets: A Comparison of NLP Techniques
    (Institute of Electrical and Electronics Engineers Inc., 2024) Balci, Emirhan; Sarac, Esra
    This paper aims to classify suicidal ideation as a symptom of depression from social media posts by applying the state-of-the-art classification model BERT (Bidirectional Encoder Representations from Transformers) and three traditional machine learning algorithms for binary classification. Since depression is one of the most prevalent mental health disorders amongst psychiatric disorders, the authors intended to present an experimental analysis of the machine learning classifier results as a comparison of novel depression detection techniques. We utilized undiagnosed user posts from Twitter as our dataset and tested the fine-tuned BERT model by applying hold-out and 10-fold cross-validation techniques. Since the dataset is highly unbalanced, Support Vector Machine (SVM), Naive Bayes, and Random Forest algorithms were employed on the same dataset with and without the oversampling method SMOTE (Synthetic Minority Oversampling Technique). The results demonstrate that traditional machine learning classifiers cannot infer sentiment from data containing various linguistic cues, such as depression symptoms. On the other hand, the state-of-the-art model BERT achieves 99.29% and 99.56% macro and micro-F-measure values, respectively, surpassing traditional machine learning algorithms in terms of these metrics. As a robust solution to depression detection from textual data, the BERT model is more trustworthy than the traditional machine learning classifiers to detect specific cues related to depression and similar mental disorders. This study contributes to the relevant research areas of natural language processing by indicating the performance difference between the BERT model and several traditional machine learning algorithms as a generalized approach for classification tasks. © 2024 IEEE.
  • [ X ]
    Öğe
    Benchmarking Large Language Model Reasoning in Indoor Robot Navigation
    (IEEE, 2025) Balci, Emirhan; Sarigul, Mehmet; Ata, Baris
    This study evaluates the performance of state-of-the-art text-based generative large language models in indoor robot navigation planning, focusing on object, spatial, and common-sense reasoning-centric instructions. Three scenes from the Matterport3D dataset were selected, along with corresponding instruction sequences and routes. Object-labeled semantic maps were generated using the RGB-D images and camera poses of the scenes. The instructions were provided to the models, and the generated robot codes were executed on a mobile robot within the selected scenes. The routes followed by the robot, which detected objects through the semantic map, were recorded. The findings indicate that while the models successfully executed object and spatial-based instructions, some models struggled with those requiring common-sense reasoning. This study aims to contribute to robotics research by providing insights into the navigation planning capabilities of language models.
  • [ X ]
    Öğe
    Effects of Extended Features on BERT Performance: Depression Detection
    (IEEE, 2025) Balci, Emirhan; Sarac, Esra
    In this study, the effects of categorical and numerical additional features obtained from Twitter posts on depression detection were investigated. Depression detection performances of the BERT large language model and SVM classifier were compared on the dataset balanced with the oversampling method. The effects of two different feature addition methods, Unimodal and Concat, were evaluated on the BERT model. The results show that oversampling improves the performance of the BERT classifier, but feature addition methods do not provide significant improvement in the model performance. The findings of the experiments reveal the success of the BERT model in the field of classification and that it does not require additional features for the detection of depression. It is believed that this study will guide research in the field of depression detection and help researchers identify more effective areas of study.