Browsing by Author "Karkiner, Zeynep"

Filter results by typing the first few letters
Now showing 1 - 2 of 2
  • No Thumbnail Available
    Item
    ParsyBot: Chatbot for Baskent University Related FAQs
    (18TH IEEE INTERNATIONAL CONFERENCE ON SEMANTIC COMPUTING, ICSC 2024, 2024-05-17) Karkiner, Zeynep; Yaman, Begum; Zengin, Begum; Cavli, Feride Nursena; Sert, Mustafa
    Reading regulations and instructions may take lots of time and sometimes it results in disappointments. To avoid this issue, people are prone to use sources that provide fast and accurate answers while accessing the information. Chatbots, are one of the most popular trend topics nowadays, and may adapted into various fields e.g., healthcare, finance, and education. This paper proposes the development of ParsyBot which is a Turkish chatbot designed to inform users about the regulations, admissions, departments, scholarships, and social clubs of Baskent University. Furthermore, users may ask via voice in Turkish this feature is not common among the other chatbots. ParsyBot uses a pre-trained BERT model which is specifically trained with regulations and instructions of Baskent University. Parsybot runs on web and mobile platforms to make it available for everyone. Our experiments on the utilized dataset, ParsyBot, reached 0.81 in METEOR, and 0.24 in ROGUE-1, which are promising compared to the ChatGPT 3.5.
  • No Thumbnail Available
    Item
    Sarcasm Detection in News Headlines with Deep Learning
    (32ND IEEE SIGNAL PROCESSING AND COMMUNICATIONS APPLICATIONS CONFERENCE, SIU 2024, 2024-12-05) Karkiner, Zeynep; Sert, Mustafa
    Sarcasm detection is one of the recent topics studied in the field of natural language processing. Although sarcasm detection is generally carried out through social media comments in the literature, it can also be applied to news headlines that are expected to be completely objective and reflect reality. In this study, sarcasm detection was carried out using various deep learning models in a dataset containing sarcastic and non-sarcastic news headlines. The accuracy of classification results of BERT, RNN, LSTM, and GRU models and their training time performance were compared. While the BERT model reached the highest accuracy (0.88), RNN was the most successful model in terms of training time performance.