http://hdl.handle.net/2117/3974 Tue, 18 Jun 2013 05:31:42 GMT 2013-06-18T05:31:42Z webmaster.bupc@upc.edu Universitat Politècnica de Catalunya. Servei de Biblioteques i Documentació no http://hdl.handle.net/2117/19484 Title: Averaging of kernel functions Authors: Belanche Muñoz, Luis Antonio; Tosi, Alessandra Abstract: In kernel-based machines, the integration of several kernels to build more flexible learning methods is a promising avenue for research. In particular, in Multiple Kernel Learning a compound kernel is build by learning a kernel that is the weighted mean of several sources. We show in this paper that the only feasible average for kernel learning is precisely the arithmetic average. We also show that three familiar means (the geometric, inverse root mean square and harmonic means) for positive real values actually generate valid kernels. Mon, 03 Jun 2013 08:59:06 GMT http://hdl.handle.net/2117/19484 2013-06-03T08:59:06Z Belanche Muñoz, Luis Antonio; Tosi, Alessandra no Kernel-based machines, Multiple kernel learning, Arithmetic average In kernel-based machines, the integration of several kernels to build more flexible learning methods is a promising avenue for research. In particular, in Multiple Kernel Learning a compound kernel is build by learning a kernel that is the weighted mean of several sources. We show in this paper that the only feasible average for kernel learning is precisely the arithmetic average. We also show that three familiar means (the geometric, inverse root mean square and harmonic means) for positive real values actually generate valid kernels. http://hdl.handle.net/2117/19468 Title: Semantics of structured normal logic programs Authors: Pasarella Sánchez, Ana Edelmira; Orejas Valdés, Fernando; Pino Blanco, Elvira; Navarro, Marisa Abstract: In this paper we provide semantics for normal logic programs enriched with structuring mechanisms and scoping rules. Specifically, we consider constructive negation and expressions of the form Q G Q in goals, where Q is a program unit, G is a goal and stands for the so-called embedded implication. Allowing the use of these expressions can be seen as adding block structuring to logic programs. In this context, we consider static and dynamic rules for visibility in blocks. In particular, we provide new semantic definitions for the class of normal logic programs with both visibility rules. For the dynamic case we follow a standard approach. We first propose an operational semantics. Then, we define a model-theoretic semantics in terms of ordered structures which are a kind of intuitionistic Beth structures. Finally, an (effective) fixpoint semantics is provided and we prove the equivalence of these three definitions. In order to deal with the static case, we first define an operational semantics and then we present an alternative semantics in terms of a transformation of the given structured programs into flat ones. We finish by showing that this transformation preserves the computed answers of the given static program. Fri, 31 May 2013 09:12:23 GMT http://hdl.handle.net/2117/19468 2013-05-31T09:12:23Z Pasarella Sánchez, Ana Edelmira; Orejas Valdés, Fernando; Pino Blanco, Elvira; Navarro, Marisa no Semantics, Normal logic programs, Embedded implication, Visibility rules, Structuring mechanism, Intuitionistic structures In this paper we provide semantics for normal logic programs enriched with structuring mechanisms and scoping rules. Specifically, we consider constructive negation and expressions of the form Q G Q in goals, where Q is a program unit, G is a goal and stands for the so-called embedded implication. Allowing the use of these expressions can be seen as adding block structuring to logic programs. In this context, we consider static and dynamic rules for visibility in blocks. In particular, we provide new semantic definitions for the class of normal logic programs with both visibility rules. For the dynamic case we follow a standard approach. We first propose an operational semantics. Then, we define a model-theoretic semantics in terms of ordered structures which are a kind of intuitionistic Beth structures. Finally, an (effective) fixpoint semantics is provided and we prove the equivalence of these three definitions. In order to deal with the static case, we first define an operational semantics and then we present an alternative semantics in terms of a transformation of the given structured programs into flat ones. We finish by showing that this transformation preserves the computed answers of the given static program. http://hdl.handle.net/2117/19457 Title: Semantically-enhanced pre-filtering for context-aware recommender systems Authors: Codina Busquet, Víctor; Ceccaroni, Luigi Abstract: Several research works have demonstrated that if users' ratings are truly context-dependent, then Context-Aware Recommender Systems can outperform traditional recommenders. In this paper we present a novel contextual pre-filtering approach that exploits the implicit semantic similarity of contextual situations. For determining such a similarity we rely only on the available users' ratings and we deem as similar two syntactically different contextual situations that are actually influencing in a similar way the user's rating behavior. We validate the proposed approach using two contextually tagged ratings data sets showing that it outperforms a traditional pre-filtering approach and a state-of-the-art context-aware Matrix Factorization model. Thu, 30 May 2013 12:29:51 GMT http://hdl.handle.net/2117/19457 2013-05-30T12:29:51Z Codina Busquet, Víctor; Ceccaroni, Luigi no Recommender systems, Behavioral research, Collaborative filtering Several research works have demonstrated that if users' ratings are truly context-dependent, then Context-Aware Recommender Systems can outperform traditional recommenders. In this paper we present a novel contextual pre-filtering approach that exploits the implicit semantic similarity of contextual situations. For determining such a similarity we rely only on the available users' ratings and we deem as similar two syntactically different contextual situations that are actually influencing in a similar way the user's rating behavior. We validate the proposed approach using two contextually tagged ratings data sets showing that it outperforms a traditional pre-filtering approach and a state-of-the-art context-aware Matrix Factorization model. http://hdl.handle.net/2117/19435 Title: A kernel for time series classification: application to atmospheric pollutants Authors: Arias Vicente, Marta; Troncoso, Alicia; Riquelme, José C. Abstract: In this paper a kernel for time-series data is presented. The main idea of the kernel is that it is designed to recognize as similar time series that may be slightly shifted with one another. Namely, it tries to focus on the shape of the time-series and ignores the fact that the series may not be perfectly aligned. The proposed kernel has been validated on several datasets based on the UCR time-series repository [1]. A comparison with the well-known Dynamic Time Warping (DTW) distance and Euclidean distance shows that the proposed kernel outperforms the Euclidean distance and is competitive with respect to the DTW distance while having a much lower computational cost. Wed, 29 May 2013 07:54:57 GMT http://hdl.handle.net/2117/19435 2013-05-29T07:54:57Z Arias Vicente, Marta; Troncoso, Alicia; Riquelme, José C. no Atmospheric pollutants, Computational costs, Data sets, Dynamic time warping, Euclidean distance, Time series classifications, Time-series data In this paper a kernel for time-series data is presented. The main idea of the kernel is that it is designed to recognize as similar time series that may be slightly shifted with one another. Namely, it tries to focus on the shape of the time-series and ignores the fact that the series may not be perfectly aligned. The proposed kernel has been validated on several datasets based on the UCR time-series repository [1]. A comparison with the well-known Dynamic Time Warping (DTW) distance and Euclidean distance shows that the proposed kernel outperforms the Euclidean distance and is competitive with respect to the DTW distance while having a much lower computational cost. http://hdl.handle.net/2117/19370 Title: Empowering automatic data-center management with machine learning Authors: Berral García, Josep Lluís; Gavaldà Mestre, Ricard; Torres Viñals, Jordi Abstract: The Cloud as computing paradigm has become nowadays crucial for most Internet business models. Managing and optimizing its performance on a moment-by-moment basis is not easy given as the amount and diversity of elements involved (hardware, applications, workloads, customer needs...). Here we show how a combination of scheduling algorithms and data mining techniques helps improving the performance and profitability of a data-center running virtualized web-services. We model the data-center's main resources (CPU, memory, IO), quality of service (viewed as response time), and workloads (incoming streams of requests) from past executions. We show how these models to help scheduling algorithms make better decisions about job and resource allocation, aiming for a balance between throughput, quality of service, and power consumption. Wed, 22 May 2013 11:19:56 GMT http://hdl.handle.net/2117/19370 2013-05-22T11:19:56Z Berral García, Josep Lluís; Gavaldà Mestre, Ricard; Torres Viñals, Jordi no The Cloud as computing paradigm has become nowadays crucial for most Internet business models. Managing and optimizing its performance on a moment-by-moment basis is not easy given as the amount and diversity of elements involved (hardware, applications, workloads, customer needs...). Here we show how a combination of scheduling algorithms and data mining techniques helps improving the performance and profitability of a data-center running virtualized web-services. We model the data-center's main resources (CPU, memory, IO), quality of service (viewed as response time), and workloads (incoming streams of requests) from past executions. We show how these models to help scheduling algorithms make better decisions about job and resource allocation, aiming for a balance between throughput, quality of service, and power consumption. http://hdl.handle.net/2117/19369 Title: Towards an implementation of a social electronic reminder for pills Authors: Gómez Sebastià, Ignasi; Garcia Gasulla, Dario; Álvarez Napagao, Sergio; Vázquez Salceda, Javier; Cortés García, Claudio Ulises Abstract: Non-compliance with prescribed medication is a major problem for elder people living alone in developed countries. Forgetfulness and confusion can lead to it, specially when multiple pathologies require a cocktail of diff erent medications each delivered at different time intervals during di fferent periods of time. Assistive technologies, a recent application area for a wide range of Arti cial Intelligence techniques and tools, have been e ectively used for supporting people in their daily activities. This paper introduces the design and implementation of a system for assisting elder people on following the treatment prescribed by a professional, with the novelty of being based in social and organisational aware assistive technology. Wed, 22 May 2013 09:00:23 GMT http://hdl.handle.net/2117/19369 2013-05-22T09:00:23Z Gómez Sebastià, Ignasi; Garcia Gasulla, Dario; Álvarez Napagao, Sergio; Vázquez Salceda, Javier; Cortés García, Claudio Ulises no Assistive technologies, Normative agents, Agent-oriented software design, Ambient intelligence Non-compliance with prescribed medication is a major problem for elder people living alone in developed countries. Forgetfulness and confusion can lead to it, specially when multiple pathologies require a cocktail of diff erent medications each delivered at different time intervals during di fferent periods of time. Assistive technologies, a recent application area for a wide range of Arti cial Intelligence techniques and tools, have been e ectively used for supporting people in their daily activities. This paper introduces the design and implementation of a system for assisting elder people on following the treatment prescribed by a professional, with the novelty of being based in social and organisational aware assistive technology. http://hdl.handle.net/2117/19368 Title: Full high-dynamic range images for dynamic scenes Authors: Ramírez Orozco, Raissel; Martín, Ignacio; Loscos, Céline; Vázquez Alcocer, Pere Pau Abstract: The limited dynamic range of digital images can be extended by composing photographs of the same scene taken with the same camera at the same view point at di erent exposure times. This is a standard procedure for static scenes but a challenging task for dynamic ones. Several methods have been presented but few recover high dynamic range within moving areas. We present a method to recover full high dynamic range (HDR) images from dynamic scenes, even in moving regions. Our method has 3 steps. Firstly, areas a ected by motion are detected to generate a ghost mask. Secondly, we register dynamic objects over a reference image (the best exposed image in the input sequence). Thirdly, we combine the registered input photographs to recover HDR values in a whole image using a weighted average function. Once matching is found, the assembling step guarantees that all aligned pixels will contribute to the nal result, including dynamic content. Tests were made on more than 20 sets of sequences, with moving cars or pedestrians and di erent background. Our results show that Image Mapping Function approach detects best motion regions while Normalized Cross Correlation o ers the best deal speed-accuracy for image registration. Results from our method o ers better result when moving object are roughly rigid and their movement is mostly rigid. The nal composition is an HDR image with no ghosting and all dynamic content present in HDR values. Wed, 22 May 2013 08:40:29 GMT http://hdl.handle.net/2117/19368 2013-05-22T08:40:29Z Ramírez Orozco, Raissel; Martín, Ignacio; Loscos, Céline; Vázquez Alcocer, Pere Pau no High dynamic range imaging, Movement detection, Image registration The limited dynamic range of digital images can be extended by composing photographs of the same scene taken with the same camera at the same view point at di erent exposure times. This is a standard procedure for static scenes but a challenging task for dynamic ones. Several methods have been presented but few recover high dynamic range within moving areas. We present a method to recover full high dynamic range (HDR) images from dynamic scenes, even in moving regions. Our method has 3 steps. Firstly, areas a ected by motion are detected to generate a ghost mask. Secondly, we register dynamic objects over a reference image (the best exposed image in the input sequence). Thirdly, we combine the registered input photographs to recover HDR values in a whole image using a weighted average function. Once matching is found, the assembling step guarantees that all aligned pixels will contribute to the nal result, including dynamic content. Tests were made on more than 20 sets of sequences, with moving cars or pedestrians and di erent background. Our results show that Image Mapping Function approach detects best motion regions while Normalized Cross Correlation o ers the best deal speed-accuracy for image registration. Results from our method o ers better result when moving object are roughly rigid and their movement is mostly rigid. The nal composition is an HDR image with no ghosting and all dynamic content present in HDR values. http://hdl.handle.net/2117/19310 Title: Cloud virtual machine scheduling: modelling the cloud virtual machine instantiation Authors: Sotiriadis, S.; Bessis, N.; Xhafa Xhafa, Fatos; Antonopoulos, Nicos Thu, 16 May 2013 13:55:24 GMT http://hdl.handle.net/2117/19310 2013-05-16T13:55:24Z Sotiriadis, S.; Bessis, N.; Xhafa Xhafa, Fatos; Antonopoulos, Nicos no Cloud, Static and dynamic virtual machine scheduling, Virtual machine instantiation, Virtualization http://hdl.handle.net/2117/19306 Title: Performance analysis of DSR and DYMO routing protocols for VANETs Authors: Spaho, Evjola; Ikeda, Makoto; Barolli, Leonard; Xhafa Xhafa, Fatos; Kolici, Vladi; Takizawa, Makoto Thu, 16 May 2013 13:36:45 GMT http://hdl.handle.net/2117/19306 2013-05-16T13:36:45Z Spaho, Evjola; Ikeda, Makoto; Barolli, Leonard; Xhafa Xhafa, Fatos; Kolici, Vladi; Takizawa, Makoto no DSR, DYMO, Routing protocols, VANETs http://hdl.handle.net/2117/19304 Title: Performance evaluation of WMN-GA system for low densities of clients and different settings of population size Authors: Oda, Tetsuya; Barolli, Admir; Spaho, Evjola; Barolli, Leonard; Xhafa Xhafa, Fatos; Takizawa, Makoto Thu, 16 May 2013 13:22:36 GMT http://hdl.handle.net/2117/19304 2013-05-16T13:22:36Z Oda, Tetsuya; Barolli, Admir; Spaho, Evjola; Barolli, Leonard; Xhafa Xhafa, Fatos; Takizawa, Makoto no Genetic algorithms, Number of covered users, Population size, Size of giant component, Wireless Mesh Networks http://hdl.handle.net/2117/19301 Title: Multimedia transmissions over a MANET testbed: problems and issues Authors: Kulla, E.; Ikeda, Makoto; Oda, Tetsuya; Barolli, Leonard; Xhafa Xhafa, Fatos; Takizawa, Makoto Thu, 16 May 2013 13:12:58 GMT http://hdl.handle.net/2117/19301 2013-05-16T13:12:58Z Kulla, E.; Ikeda, Makoto; Oda, Tetsuya; Barolli, Leonard; Xhafa Xhafa, Fatos; Takizawa, Makoto no Audio, BATMAN, MANET testbed, Multimedia transmissions, Problems and issues, Video http://hdl.handle.net/2117/19300 Title: Data replication and synchronization in P2P collaborative systems Authors: Xhafa Xhafa, Fatos Thu, 16 May 2013 13:04:29 GMT http://hdl.handle.net/2117/19300 2013-05-16T13:04:29Z Xhafa Xhafa, Fatos no Data replication, data synchronization, P2P systems, groupware http://hdl.handle.net/2117/19264 Title: A comparison study between two fuzzy-based trustworthiness systems for P2P networks Authors: Umezaki, K.; Spaho, Evjola; Barolli, Leonard; Xhafa Xhafa, Fatos; Matsuo, Keita; Iwashige, J. Wed, 15 May 2013 15:47:19 GMT http://hdl.handle.net/2117/19264 2013-05-15T15:47:19Z Umezaki, K.; Spaho, Evjola; Barolli, Leonard; Xhafa Xhafa, Fatos; Matsuo, Keita; Iwashige, J. no Fuzzy system, JXTA-Overlay, P2P, Trustworthiness http://hdl.handle.net/2117/19263 Title: Tuning of operators in memetic algorithms for independent batch scheduling in computational grids Authors: Xhafa Xhafa, Fatos; Duran, B.; Barolli, Leonard; Kolici, Vladi; Miho, Rozeta; Takizawa, Makoto Wed, 15 May 2013 15:34:03 GMT http://hdl.handle.net/2117/19263 2013-05-15T15:34:03Z Xhafa Xhafa, Fatos; Duran, B.; Barolli, Leonard; Kolici, Vladi; Miho, Rozeta; Takizawa, Makoto no Computational Grids, Memetic Algorithms, Meta-heuristics, Operators, Tuning http://hdl.handle.net/2117/19260 Title: Mission operations scheduling: complexity and resolution methods Authors: Xhafa Xhafa, Fatos; Sun, Junzi; Barolli, Admir; Takizawa, Makoto Wed, 15 May 2013 15:05:24 GMT http://hdl.handle.net/2117/19260 2013-05-15T15:05:24Z Xhafa Xhafa, Fatos; Sun, Junzi; Barolli, Admir; Takizawa, Makoto no Constraint programming, Genetic Algorithms, Ground station scheduling, Satellite scheduling, Simulation