3 sonuçlar
Arama Sonuçları
Listeleniyor 1 - 3 / 3
Yayın Driver recognition using gaussian mixture models and decision fusion techniques(Springer-Verlag Berlin, 2008) Benli, Kristin Surpuhi; Düzağaç, Remzi; Eskil, Mustafa TanerIn this paper we present our research in driver recognition. The goal of this study is to investigate the performance of different classifier fusion techniques in a driver recognition scenario. We are using solely driving behavior signals such as break and accelerator pedal pressure, engine RPM, vehicle speed; steering wheel angle for identifying the driver identities. We modeled each driver using Gaussian Mixture Models, obtained posterior probabilities of identities and combined these scores using different fixed mid trainable (adaptive) fusion methods. We observed error rates is low as 0.35% in recognition of 100 drivers using trainable combiners. We conclude that the fusion of multi-modal classifier results is very successful in biometric recognition of a person in a car setting.Yayın Improving search engine performance with context extraction using Lucene, DBPedia-Spotlight, and Wordnet(Işık Üniversitesi, 2014) Düzağaç, Remzi; Yıldız, Olcay Taner; Işık Üniversitesi, Fen Bilimleri Enstitüsü, Bilgisayar Mühendisliği Yüksek Lisans ProgramıSearch engines are common tools which retrieve information from considerable amount of data according to the user needs. The data size that needs to be handled and retrieving relevant information, are the main problems of every search engine. Additionally, in order to improve the performance of a search engine, there are various approaches and methods are applied. On the other hand, using context information besides words in the document is a quite new area. Including Context Information" into the game is a promising eld of work. In this research, we use context information extracted from the documents in the collection to improve the performance of the search engine. In rst step, we extract context using Lucene, DBPedia-Spotlight, and Wordnet. As the second step, we build a graph using extracted context information. In the third step, in order to group similar contexts, we cluster context graph. In the fourth step, we rescore results using context-clusters and context-information of documents, as well as queries. In the fth step, we implement a data collection tool to collect gold-standard data. In the sixth and nal step, we compare the results of our algorithm with gold-standard data set. According to experimental results, using context information may improve the search engine performance but the collection should be relatively big.Yayın Context sensitive search engine(Springer-Verlag Berlin, 2014-10-27) Düzağaç, Remzi; Yıldız, Olcay TanerIn this paper, we use context information extracted from the documents in the collection to improve the performance of the search engine. In first step, we extract context using Lucene, DBPedia-Spotlight, and Wordnet. As the second step, we build a graph using extracted context information. In the third step, in order to group similar contexts, we cluster context graph. In the fourth step, we re-score results using context-clusters and context-information of documents, as well as queries. In the fifth step, we implement a data collection tool to collect gold-standard data. In the sixth and final step, we compare the results of our algorithm with gold-standard data set. According to the experimental results, using context information may improve the search engine performance but the collection should be relatively big.
