3 sonuçlar
Arama Sonuçları
Listeleniyor 1 - 3 / 3
Yayın An experimental evaluation of prior polarities in sentiment lexicons(IEEE, 2017) Kanburoğlu, Ali Buğra; Solak, ErcanWe present the results of an experiment to assess the validity of prior polarities available in sentiment lexicons. We designed a ranking task that was elicited through pairwise comparisons and compared the results to those predicted by two popular sentiment lexicons. We find that the experiment results show a moderate level of agreement between the lexicons and human judgments.Yayın Multilingual information retrieval on the Internet: A case study of Turkish users(Academic Press Ltd- Elsevier Science Ltd, 2005-12) Aytaç, SelenayThis study aims to answer the following research question: What information retrieval problems do Turkish Internet users face by using Turkish on the Internet?The data for this report were gathered by triangulation of three different methods: (1) e-mail questionnaire survey, (2) face-to-face interviews, and (3) participant observation of Turkish speaking respondents, in order to assess the major obstacles of retrieving Turkish language information by using Turkish on the Internet. Although a significant amount of research has been focused on multilingual information retrieval, a review of the literature reveals that this pilot study is the first initiative to draw a picture from the Turkish Internet user's point of view.Yayın Privacy-preserving cyber threat intelligence: a framework combining private information retrieval, federated learning, and differential privacy(Institute of Electrical and Electronics Engineers Inc., 2025-09-21) Çamalan, Emre; Çeliktaş, BarışThreat Intelligence Platforms (TIPs) are essential for sharing indicators of compromise (IoCs), but querying them can leak sensitive organizational data. We propose a privacy-preserving framework that combines Private Information Retrieval (PIR), Federated Learning (FL), and Differential Privacy (DP) to mitigate this risk. Our approach addresses both content-level and metadata-level privacy concerns while supporting collaborative learning across organizations. It ensures that sensitive query patterns remain hidden, local threat data never leaves organizational boundaries, and model updates are protected against inference attacks. The framework integrates with existing TIPs such as MISP and OpenCTI, requiring minimal operational changes. We implement a prototype using a simulated Abuse IP dataset and evaluate it on latency, accuracy, and communication overhead. The system supports private queries in under 300 ms and maintains over 95% model accuracy under DP noise. These results indicate that strong privacy can be achieved with minimal performance trade-offs, making the approach viable for real-world CTI environments.
