Recommender System (RS) predicts user's ratings towards items, and then recommends highly-predicted items to user. In recent years, RS has been playing more and more important role in the agent research field. There have been a great deal of researches trying to apply agent technology to RS. Collaborative Filtering, one of the most widely used approach to predict user's ratings in Recommender System, predicts a user's rating towards an item by aggregating ratings given by users who have similar preference to that user. In existing approaches, user similarity is often computed on the whole set of items. However, because the number of items is often very large and so is the diversity among items, users who have similar preference in one category may have totally different judgement on items of another kind. In order to deal with this problem, we propose a method to cluster items, so that inside a cluster, similarity between users does not change significantly from item to item. After the item clustering phase, when predicting rating of a user towards an item, we only aggregate ratings of users who have similarity preference to that user inside the cluster of that item. Experiments evaluating our approach are carried out on the real dataset taken from MovieLens, a movies recommendation web site. Experiment results suggest that our approach can improve prediction accuracy compared to existing approaches.

For a microstrip antenna (MSA) with a ring-shaped slot on formed on the ground plane, downsizing the microstrip patch and expanding the circularly polarized bandwidth have been achieved successfully. The dimensions of the patch are 6.8 mm7.4 mm and the minimum axial ratio (AR) of 0.6 dB is obtained at 6.1 GHz. In addition, its AR is less than 3 dB at the relative bandwidth of 3.5%. The bandwidth of the proposed MSA is twice that of conventional single-feeding circularly polarized MSAs; however, its size is only half that of conventional MSAs.

In this paper, a simple piezoelectric floating mass transducer (PFMT) for implantable middle ear hearing devices (IMEHDs) is proposed and its modeling and designing are studied. The transducer which can be implanted in the meddle ear consists of a PMN-PT multi-layered piezoelectric actuator, an elastic material, and a metal case. The proposed transducer has a simple structure and the force generated from the piezoelectric actuator is efficiently transferred to the ossicles of the middle ear. For the analysis of the vibration characteristics, the transducer attached on the ossicle is simplified into a simple mechanical model considering the mass of an incus. And the vibration displacement of the model is calculated using computer simulation and verified by the experimental results. It is shown that the designed PFMT can allow implantation in the middle ear cavity and provide a sufficiently high output of more than 100 nm of vibration displacement. Plus, it is verified that the vibration characteristics of PFMT can be controlled through adjustment of the metal case size and the elastic material of the transducer.

We survey traffic matrix models, whose elements represent the traffic demand between source-destination pair nodes. Modeling the traffic matrix is useful for multilayer Traffic Engineering (TE) in IP optical networks. Multilayer TE techniques make the network so designed flexible and reliable. This is because it allows reconfiguration of the virtual network topology (VNT), which consists of a set of several lower-layer (optical) paths and is provided to the higher layer, in response to fluctuations (diurnal) in traffic demand. It is, therefore, important to synthetically generate traffic matrices as close to the real ones as possible to maximize the performance of multilayer TE. We compare several models and clarify their applicability to VNT design and control. We find that it is difficult in practice to make an accurate traffic matrix with conventional schemes because of the high cost for data measurement and the complicated calculations involved. To overcome these problems, we newly introduce a simplified traffic matrix model that is practical; it well mirrors real networks. Next, this paper presents our developed server, the IP Optical TE server. It performs multilayer TE in IP optical networks. We evaluate the effectiveness of multilayer TE using our developed IP Optical server and the simplified traffic matrix. We confirm that multilayer TE offers significant CAPEX savings. Similarly, we demonstrate basic traffic control in IP optical networks, and confirm the dynamic control of the network and the feasibility of the IP Optical TE server.

In this paper, a new method for clustering of players in order to analyze games in soccer videos is proposed. The proposed method classifies players who are closely related in terms of soccer tactics into one group. Considering soccer tactics, the players in one group are located near each other. For this reason, the Euclidean distance between the players is an effective measurement for the clustering of players. However, the distance is not sufficient to extract tactics-based groups. Therefore, we utilize a modified version of the community extraction method, which finds community structure by dividing a non-directed graph. The use of this method in addition to the distance enables accurate clustering of players.

In order to realize a higher density version of the conventional optical disk, shorter wavelength laser and narrower track pitch have been put to practical use. However, using narrow track pitch can cause the increase of the crosstalk from the adjacent tracks. Moreover, the use of narrow pitch and short wavelength can also give rise to the increase of deterioration of the detected signal characteristics due to the microscopic roughness of disk surface. In this paper, in order to estimate the effect of surface roughness theoretically, we try to analyze the light-beam scattering and detected signal characteristics of a blue laser optical disk model with random rough surfaces by the Finite Difference Time Domain (FDTD) method.

Digital hearing aid users often complain of difficulty in understanding speech in the presence of background noise. To improve speech perception in a noisy environment, various speech enhancement algorithms have been applied in digital hearing aids. In this study, a speech enhancement algorithm using modified spectral subtraction and companding is proposed for digital hearing aids. We adjusted the biases of the estimated noise spectrum, based on a subtraction factor, to decrease the residual noise. Companding was applied to the channel of the formant frequency based on the speech presence indicator to enhance the formant. Noise suppression was achieved while retaining weak speech components and avoiding the residual noise phenomena. Objective and subjective evaluation under various environmental conditions confirmed the improvement due to the proposed algorithm. We tested segmental SNR and Log Likelihood Ratio (LLR), which have higher correlation with subjective measures. Segmental SNR has the highest and LLR the lowest correlation of the methods tested. In addition, we confirmed by spectrogram that the proposed method significantly reduced the residual noise and enhanced the formants. A mean opinion score that represented the global perception score was tested; this produced the highest quality speech using the proposed method. The results show that the proposed speech enhancement algorithm is beneficial for hearing aid users in noisy environments.

The specification of a Multi-Agent System (MAS) involves the identification of a large number of entities and their relationships. This is a non-trivial task that requires managing different views of the system. Many problems concerning this issue originate in the presence of contradictory goals and tasks, inconsistencies, and unexpected behaviours. Such troublesome configurations should be detected and prevented during the development process in order to study alternative ways to cope with them. In this paper, we present methods and tools that support the management of contradictions during the analysis and design of MAS. Contradiction management in MAS has to consider both individual (i.e. agent) and social (i.e. organization) aspects, and their dynamics. Such issues have already been considered in social sciences, and more concretely in the Activity Theory, a social framework for the study of interactions in activity systems. Our approach applies knowledge from Activity Theory in MAS, especially its base of contradiction patterns. That requires a formalization of this social theory in order to be applicable in a software engineering context and its adaptation to agent-oriented methodologies. Then, it will be possible to check the occurrence of contradiction patterns in a MAS specification and provide solutions to those situations. This technique has been validated by implementing an assistant for the INGENIAS Development Kit and has been tested with several case studies. This paper shows part of one of these experiments for a web application.

This paper considers optical transport and packet networks and discusses the constraints and solutions in computation of traffic engineering paths. We categorize the constraints into prunable or non-prunable classes. The former involves a simple metric which can be applied for filtering to determine the path. The latter requires a methodic consideration of more complicated network element attributes. An example of this type of constraints is path loss in which the metric can be evaluated only on a path basis, as opposed to simply applying the metric to the link. Another form of non-prunable constraint requires adaptation and common vector operation. Examples are the switching type adaptation and wavelength continuity, respectively. We provide possible solutions to cases with different classes of constraints and address the problem of path computation in support of traffic engineering in multi-layer networks where a set of constrains are concurrently present. The solutions include the application of channel graph and common vector to support switching type adaptation and label continuity, respectively.

The article describes recent adaptive estimation algorithms over distributed networks. The algorithms rely on local collaborations and exploit the space-time structure of the data. Each node is allowed to communicate with its neighbors in order to exploit the spatial dimension, while it also evolves locally to account for the time dimension. Algorithms of the least-mean-squares and least-squares types are described. Both incremental and diffusion strategies are considered.

A novel type Brillouin optical time-domain analysis (BOTDA), called double-pulse BOTDA (DP-BOTDA), is proposed for measuring distributed strain and temperature in a fiber with a centimeter spatial resolution. The DP-BOTDA system transmits a double-pulsed light instead of a conventional single-pulsed light into a fiber to interact with a counter-propagating continuous-wave light through the induced acoustic wave in the fiber. The interference between acoustic waves induced by the front and rear pulses of the double-pulsed light produces broad but oscillatory Brillouin gain spectra that make it possible to measure the Brillouin frequency shift accurately despite the very narrow pulse width. Our numerical simulation, which includes an estimation of the signal-to-noise ratio of the system, shows that it is possible to measure the distributed Brillouin frequency shift with a spatial resolution of 4 cm and accuracies of 1-2 MHz for a 5-km long fiber.

The present paper proposes two novel and practical schemes for distributed and asynchronous power control in wireless ad hoc networks, in which users dynamically share several frequency bands as in "cognitive radio" networks. These schemes iteratively adjust transmit powers of individual network transmitters with respect to mutually caused interference in the shared bands. Their most attractive feature is that they find network-wide acceptable trade-offs to diverse signal-to-noise and interference (SINR) requirements and efficiently use techniques of stochastic approximation and time-averaging to guarantee a robust performance in random channels. Advantageously, both proposed algorithms do not assume any particular modulation, coding, QoS measure definition or network architecture, which assures their high applicability in the industry and research. Moreover, the broad definition and non-linear nature of these schemes mathematically generalize and thus encompass as a special case many widely deployed power control schemes such as e.g. those for achieving fixed SINR targets or using game-theoretic utility maximization. Simulations are provided to illustrate our approach and its better performance compared to standard algorithms.

Internet access traffic follows hourly patterns that depend on various factors, such as the periods users stay on-line at the access point (e.g. at home or in the office) or their preferences for applications. The clustering of Internet users may provide important information for traffic engineering and billing. For example, it can be used to set up service differentiation according to hourly behavior, resource optimization based on multi-hour routing and definition of tariffs that promote Internet access in low busy hours. In this work, we propose a methodology for clustering Internet users with similar patterns of Internet utilization, according to their hourly traffic utilization. The methodology resorts to three statistical multivariate analysis techniques: cluster analysis, principal component analysis and discriminant analysis. The methodology is illustrated through measured data from two distinct ISPs, one using a CATV access network and the other an ADSL one, offering distinct traffic contracts. Principal component analysis is used as an exploratory tool. Cluster analysis is used to identify the relevant Internet usage profiles, with the partitioning around medoids and Ward's method being the preferred clustering methods. For the two data sets, these methods lead to the choice of 3 clusters with different hourly traffic utilization profiles. The cluster structure is validated through discriminant analysis. It is also evaluated in terms of several characteristics of the user traffic not used in the cluster analysis, such as the type of applications, the amount of downloaded traffic, the activity duration and the transfer rate, resulting in coherent outcomes.

This paper investigates the use of the affine transformation matrix when employing principal component analysis (PCA) to compress the data of 3D animation models. Satisfactory results were achieved for the common 3D models by using PCA because it can simplify several related variables to a few independent main factors, in addition to making the animation identical to the original by using linear combinations. The selection of the principal component factor (also known as the base) is still a subject for further research. Selecting a large number of bases could improve the precision of the animation and reduce distortion for a large data volume. Hence, a formula is required for base selection. This study develops an automatic PCA selection method, which includes the selection of suitable bases and a PCA separately on the three axes to select the number of suitable bases for each axis. PCA is more suitable for animation models for apparent stationary movement. If the original animation model is integrated with transformation movements such as translation, rotation, and scaling (RTS), the resulting animation model will have a greater distortion in the case of the same base vector with regard to apparent stationary movement. This paper is the first to extract the model movement characteristics using the affine transformation matrix and then to compress 3D animation using PCA. The affine transformation matrix can record the changes in the geometric transformation by using 44 matrices. The transformed model can eliminate the influences of geometric transformations with the animation model normalized to a limited space. Subsequently, by using PCA, the most suitable base vector (variance) can be selected more precisely.

This letter presents an improved visible watermarking scheme for halftone images. It incorporates watermark embedding into ordered dither halftoning by threshold modulation. The input images include a continuous-tone host image (e.g. an 8-bit gray level image) and a binary watermark image, and the output is a halftone image with a visible watermark. Our method is content adaptive because it takes local intensity information of the host image into account. Experimental results demonstrate effectiveness of the proposed technique. It can be used in practical applications for halftone images, such as commercial advertisement, content annotation, copyright announcement, etc.

In ad-hoc networks, mobile nodes are limited by a range of radio coverage and have an irregular source of power due to their battery. In ad-hoc networks, there are a lot of situations that all mobile nodes need to agree on their key not at the same time but in part and then merge themselves subsequently. This is because ad-hoc networks have specific features such as mobility and allow various conditions during configuration. In this thesis, we propose MCP (Merging Clusters Protocol), a simple key agreement scheme that can effectively deal with merging different adjacent clusters in mobile ad-hoc networks. When nodes of each cluster have already agreed on their own group keys and intend to merge themselves for further secure communications, MCP can be used in an efficient and secure way. In addition, it can be utilized for efficient group key agreement in a large ad-hoc network. We analyze the security and efficiency of MCP and discuss the experimental results according to practical implementation scenarios.

We propose a recursive clustering and order restriction (R-CORE) method for learning large-scale Bayesian networks. The proposed method considers a reduced search space for directed acyclic graph (DAG) structures in scoring-based Bayesian network learning. The candidate DAG structures are restricted by clustering variables and determining the intercluster directionality. The proposed method considers cycles on only cmax(«n) variables rather than on all n variables for DAG structures. The R-CORE method could be a useful tool in very large problems where only a very small amount of training data is available.

This paper presents a novel feature extraction algorithm based on particle swarms for processing hyperspectral imagery data. Particle swarm optimization, originally developed for global optimization over continuous spaces, is extended to deal with the problem of feature extraction. A formulation utilizing two swarms of particles was developed to optimize simultaneously a desired performance criterion and the number of selected features. Candidate feature sets were evaluated on a regression problem. Artificial neural networks were trained to construct linear and nonlinear models of chemical concentration of glucose in soybean crops. Experimental results utilizing real-world hyperspectral datasets demonstrate the viability of the method. The particle swarms-based approach presented superior performance in comparison with conventional feature extraction methods, on both linear and nonlinear models.

We propose a robust iterative multiuser receiver for decoding convolutional coded code-division multiple access (CDMA) signals in both Gaussian and non-Gaussian channel noise. The receiver is derived from a modified maximum a-posteriori (MAP) algorithm called the max-log-MAP algorithm for robustness against erroneous channel variance estimation. Furthermore, the effect of destructive outliers arising from impulsive noise is mitigated in the proposed receiver by incorporating the robust Huber penalty function into the multiuser detector. The proposed receiver is shown to perform satisfactorily over Gaussian and non-Gaussian impulsive channels. In every iteration, cumulative improvement in the quality of the a-posteriori probabilities is also demonstrated.

In this paper, we introduce a syntactically embedded (s-embedded) language, and consider its principal congruence. The following three results are proved, where PL is the principal congruence of a language L, and W(L) is the residual of L. (1) For a language K, s-embedded in M, K is equal to a PM class. (2) For a language K, s-embedded in an infix language M, K is equal to a PW(M) class. (3) For a nonempty s-embedded language L, if L is double-unitary, then L is equal to a PW(M) class. From the above results, we can obtain those for principal congruence of some codes. For example, Ln is equal to a PLn+1 class for an inter code L of index n.