Özgür Ulusoy

Bio: Özgür Ulusoy is an academic researcher from Bilkent University. The author has contributed to research in topics: Web query classification & Web search query. The author has an hindex of 32, co-authored 154 publications receiving 3548 citations. Previous affiliations of Özgür Ulusoy include University of Illinois at Urbana–Champaign.

A data mining approach for location prediction in mobile environments

Abstract: Mobility prediction is one of the most essential issues that need to be explored for mobility management in mobile computing systems. In this paper, we propose a new algorithm for predicting the next intercell movement of a mobile user in a Personal Communication Systems network. In the first phase of our three-phase algorithm, user mobility patterns are mined from the history of mobile user trajectories. In the second phase, mobility rules are extracted from these patterns, and in the last phase, mobility predictions are accomplished by using these rules. The performance of the proposed algorithm is evaluated through simulation as compared to two other prediction methods. The performance results obtained in terms of Precision and Recall indicate that our method can make more accurate predictions than the other methods.

A framework for use of wireless sensor networks in forest fire detection and monitoring

Abstract: Forest fires are one of the main causes of environmental degradation nowadays. Current surveillance systems for forest fires lack in supporting real-time monitoring of every point of a region at all times and early detection of fire threats. Solutions using wireless sensor networks, on the other hand, can gather sensory data values, such as temperature and humidity, from all points of a field continuously, day and night, and, provide fresh and accurate data to the fire-fighting center quickly. However, sensor networks face serious obstacles like limited energy resources and high vulnerability to harsh environmental conditions, that have to be considered carefully. In this paper, we propose a comprehensive framework for the use of wireless sensor networks for forest fire detection and monitoring. Our framework includes proposals for the wireless sensor network architecture, sensor deployment scheme, and clustering and communication protocols. The aim of the framework is to detect a fire threat as early as possible and yet consider the energy consumption of the sensor nodes and the environmental conditions that may affect the required activity level of the network. We implemented a simulator to validate and evaluate our proposed framework. Through extensive simulation experiments, we show that our framework can provide fast reaction to forest fires while also consuming energy efficiently.

A quadtree-based dynamic attribute indexing method

Abstract: Dynamic attributes are attributes that change continuously over time making it impractical to issue explicit updates for every change. In this paper, we adapt a variant of the quadtree structure to solve the problem of indexing dynamic attributes. The approach is based on the key idea of using a linear function of time for each dynamic attribute that allows us to predict its value in the future. We contribute an algorithm for regenerating the quadtree-based index periodically that minimizes CPU and disk access cost. We also provide an experimental study of performance focusing on query processing and index update overheads.

Free Riding in Peer-to-Peer Networks

Abstract: Free riding in peer-to-peer (P2P) networks poses a serious threat to their proper operation. Here, the authors present a variety of approaches developed to overcome this problem. They introduce several unique aspects of P2P networks and discuss free riding's effects on P2P services. They categorize proposed solutions and describe each category's important features and implementation issues together with some sample solutions. They also discuss open issues, including common attacks and security considerations.

A Transaction Model for Multidatabase Systems

Abstract: In this paper, we present a new transaction model for multidatabase systems (MDBSs). This model captures the formalism and semantics of various extended transaction models and adopts them to an MDBS environment. The extended models constituting our transaction model are the nested transactions [4], the flexible transaction model that provides various dependency relations among transactions [7], and the model that involves a relaxed version of transaction atomicity, namely the semantic atomicity, to increase the level of concurrency [2], [5]. While including the semantics of all those transaction models, the global serializability in our execution model was ensured through the use of the ticketing method [3].

When is nearest neighbor meaningful

Abstract: We explore the effect of dimensionality on the nearest neighbor problem. We show that under a broad set of conditions (much broader than independent and identically distributed dimensions), as dimensionality increases, the distance to the nearest data point approaches the distance to the farthest data point. To provide a practical perspective, we present empirical results on both real and synthetic data sets that demonstrate that this effect can occur for as few as 10-15 dimensions. These results should not be interpreted to mean that high-dimensional indexing is never meaningful; we illustrate this point by identifying some high-dimensional workloads for which this effect does not occur. However, our results do emphasize that the methodology used almost universally in the database literature to evaluate high-dimensional indexing techniques is flawed, and should be modified. In particular, most such techniques proposed in the literature are not evaluated versus simple linear scan, and are evaluated over workloads for which nearest neighbor is not meaningful. Often, even the reported experiments, when analyzed carefully, show that linear scan would outperform the techniques being proposed on the workloads studied in high (10-15) dimensionality!.

Multimodal learning with deep Boltzmann machines

Abstract: Data often consists of multiple diverse modalities For example, images are tagged with textual information and videos are accompanied by audio Each modality is characterized by having distinct statistical properties We propose a Deep Boltzmann Machine for learning a generative model of such multimodal data We show that the model can be used to create fused representations by combining features across modalities These learned representations are useful for classification and information retrieval By sampling from the conditional distributions over each data modality, it is possible to create these representations even when some data modalities are missing We conduct experiments on bimodal image-text and audio-video data The fused representation achieves good classification results on the MIR-Flickr data set matching or outperforming other deep models as well as SVM based models that use Multiple Kernel Learning We further demonstrate that this multimodal model helps classification and retrieval even when only unimodal data is available at test time

Multimodal Learning with Deep Boltzmann Machines

Abstract: A Deep Boltzmann Machine is described for learning a generative model of data that consists of multiple and diverse input modalities. The model can be used to extract a unified representation that fuses modalities together. We find that this representation is useful for classification and information retrieval tasks. The model works by learning a probability density over the space of multimodal inputs. It uses states of latent variables as representations of the input. The model can extract this representation even when some modalities are absent by sampling from the conditional distribution over them and filling them in. Our experimental results on bi-modal data consisting of images and text show that the Multimodal DBM can learn a good generative model of the joint space of image and text inputs that is useful for information retrieval from both unimodal and multimodal queries. We further demonstrate that this model significantly outperforms SVMs and LDA on discriminative tasks. Finally, we compare our model to other deep learning methods, including autoencoders and deep belief networks, and show that it achieves noticeable gains.

Indexing the positions of continuously moving objects

Abstract: The coming years will witness dramatic advances in wireless communications as well as positioning technologies. As a result, tracking the changing positions of objects capable of continuous movement is becoming increasingly feasible and necessary. The present paper proposes a novel, R*-tree based indexing technique that supports the efficient querying of the current and projected future positions of such moving objects. The technique is capable of indexing objects moving in one-, two-, and three-dimensional space. Update algorithms enable the index to accommodate a dynamic data set, where objects may appear and disappear, and where changes occur in the anticipated positions of existing objects. A comprehensive performance study is reported.