There are increasing demands for improved analysis of multimodal data that consist of multiple representations, such as multilingual documents and text-annotated images. One promising approach for analyzing such multimodal data is latent topic models. In this paper, we propose conditionally independent generalized relational topic models (CI-gRTM) for predicting unknown relations across different multiple representations of multimodal data. We developed CI-gRTM as a multimodal extension of discriminative relational topic models called generalized relational topic models (gRTM). We demonstrated through experiments with multilingual documents that CI-gRTM can more effectively predict both multilingual representations and relations between two different language representations compared with several state-of-the-art baseline models that enable to predict either multilingual representations or unimodal relations.

In Twitter-like services, countless messages are being posted in real-time every second all around the world. Timely knowledge about what kinds of information are diffusing in social media is quite important. For example, in emergency situations such as earthquakes, users provide instant information on their situation through social media. The collective intelligence of social media is useful as a means of information detection complementary to conventional observation. We have developed a system for monitoring and analyzing information diffusion data in real-time by tracking retweeted tweets. A tweet retweeted by many users indicates that they find the content interesting and impactful. Analysts who use this system can find tweets retweeted by many users and identify the key people who are retweeted frequently by many users or who have retweeted tweets about particular topics. However, bursting situations occur when thousands of social media messages are suddenly posted simultaneously, and the lack of machine resources to handle such situations lowers the system's query performance. Since our system is designed to be used interactively in real-time by many analysts, waiting more than one second for a query results is simply not acceptable. To maintain an acceptable query performance, we propose a capacity control method for filtering incoming tweets using extra attribute information from tweets themselves. Conventionally, there is a trade-off between the query performance and the accuracy of the analysis results. We show that the query performance is improved by our proposed method and that our method is better than the existing methods in terms of maintaining query accuracy.

Load size for a service on the Internet changes remarkably every hour. Thus, it is expected for service system scales to change dynamically according to load size. KVS (key-value store) is a scalable DBMS (database management system) widely used in largescale Internet services. In this paper, we focus on Cassandra, a popular open-source KVS implementation, and discuss methods for improving dynamic scaling performance. First, we evaluate node joining time, which is the time to complete adding a node to a running KVS system, and show that its bottleneck process is disk I/O. Second, we analyze disk accesses in the nodes and indicate that some heavily accessed files cause a large number of disk accesses. Third, we propose two methods for improving elasticity, which means decreasing node adding and removing time, of Cassandra. One method reduces disk accesses significantly by keeping the heavily accessed file in the page cache. The other method optimizes I/O scheduler behavior. Lastly, we evaluate elasticity of our methods. Our experimental results demonstrate that the methods can improve the scaling-up and scaling-down performance of Cassandra.

Microblog, especially twitter, has become an integral part of our daily life for searching latest news and events information. Due to the short length characteristics of tweets and frequent use of unconventional abbreviations, content-relevance based search cannot satisfy user's information need. Recent research has shown that considering temporal and contextual aspects in this regard has improved the retrieval performance significantly. In this paper, we focus on microblog retrieval, emphasizing the alleviation of the vocabulary mismatch, and the leverage of the temporal (e.g., recency and burst nature) and contextual characteristics of tweets. To address the temporal and contextual aspect of tweets, we propose new features based on query-tweet time, word embedding, and query-tweet sentiment correlation. We also introduce some popularity features to estimate the importance of a tweet. A three-stage query expansion technique is applied to improve the relevancy of tweets. Moreover, to determine the temporal and sentiment sensitivity of a query, we introduce query type determination techniques. After supervised feature selection, we apply random forest as a feature ranking method to estimate the importance of selected features. Then, we make use of ensemble of learning to rank (L2R) framework to estimate the relevance of query-tweet pair. We conducted experiments on TREC Microblog 2011 and 2012 test collections over the TREC Tweets2011 corpus. Experimental results demonstrate the effectiveness of our method over the baseline and known related works in terms of precision at 30 (P@30), mean average precision (MAP), normalized discounted cumulative gain at 30 (NDCG@30), and R-precision (R-Prec) metrics.

Recurrent neural networks (RNNs) are a powerful model for sequential data. RNNs that use long short-term memory (LSTM) cells have proven effective in handwriting recognition, language modeling, speech recognition, and language comprehension tasks. In this study, we propose LSTM conditional random fields (LSTM-CRF); it is an LSTM-based RNN model that uses output-label dependencies with transition features and a CRF-like sequence-level objective function. We also propose variations to the LSTM-CRF model using a gate recurrent unit (GRU) and structurally constrained recurrent network (SCRN). Empirical results reveal that our proposed models attain state-of-the-art performance for named entity recognition.

This paper presents an unsupervised approach to feature binary coding for efficient semantic image retrieval. Although the majority of the existing methods aim to preserve neighborhood structures of the feature space, semantically similar images are not always in such neighbors but are rather distributed in non-linear low-dimensional manifolds. Moreover, images are rarely alone on the Internet and are often surrounded by text data such as tags, attributes, and captions, which tend to carry rich semantic information about the images. On the basis of these observations, the approach presented in this paper aims at learning binary codes for semantic image retrieval using multimodal information sources while preserving the essential low-dimensional structures of the data distributions in the Hamming space. Specifically, after finding the low-dimensional structures of the data by using an unsupervised sparse coding technique, our approach learns a set of linear projections for binary coding by solving an optimization problem which is designed to jointly preserve the extracted data structures and multimodal data correlations between images and texts in the Hamming space as much as possible. We show that the joint optimization problem can readily be transformed into a generalized eigenproblem that can be efficiently solved. Extensive experiments demonstrate that our method yields significant performance gains over several existing methods.

We propose a Simulink model of a ring oscillator using saturating integrators. The oscillator's period is tuned via the saturation time of the integrators. Thus, timing jitters due to white and flicker noises are easily introduced into the model, enabling an efficient phase noise evaluation before transistor-level circuit design.

In this paper we study the structure of self-dual cyclic codes over the ring $Lambda= Z_4+uZ_4$. The ring Λ is a local Frobenius ring but not a chain ring. We characterize self-dual cyclic codes of odd length n over Λ. The results can be used to construct some optimal binary, quaternary cyclic and self-dual codes.

We propose a pre-filtering system for blind equalization in order to separate orthogonal frequency division multiplexing (OFDM) symbols in a multiple-input multiple-output (MIMO) - OFDM system. In a conventional blind MIMO-OFDM equalization without the pre-filtering system, there is a possibility that originally transmitted streams are permutated, resulting in the receiver being unable to retrieve desired signals. We also note that signal permutation is different for each subcarrier. In order to solve this problem, each transmitted stream of the proposed MIMO-OFDM system is pre-filtered by a unique allpass filter. In this paper, the pre-filter is referred to as transmit tagging filter (TT-Filter). At a receiver, an inverse filter of the TT-filter is used to blindly equalize a MIMO channel without permutation problem. Further, in order to overcome the issue of phase ambiguity, this paper introduces blind phase compensation.

In this paper, we investigate the channel characteristics of underwater optical wireless communications (UOWC) based on Monte Carlo simulation method. The impulse response and channel time dispersion of the link are discussed. Also we consider the channel parameters comprehensively like the water type, attenuation length, divergence angle, beam width, field-of-view (FOV), receiver aperture and position. Simulation results suggest that in clear water, the channel can effectively be considered as non inter-symbol interference (ISI) when working over distance of up to 40m. Therefore, in practice the receiver does not need to perform computationally complex signal processing operations. However, in harbor water, the channel time dispersion will enlarge with larger FOV or divergence angle, and reduce the data transmission efficiency. When the attenuation length is smaller than diffused length, larger receivers offer lower intensity than smaller ones. In contrast, the intensity enhances with larger receiver at the small FOV, however, they trend to similar regardless of the apertures at large FOV. Furthermore, we study the effect of misalignment of the transmitter and receiver on the received intensity. The results give us some insight in terms of what constitutes an accurate UOWC channel.

In this paper, a Boolean algebra approach is proposed to encode various acceptability semantics for abstract argumentation frameworks, where each semantics can be equivalently encoded into several Boolean constraint models based on Boolean matrices and a family of Boolean operations between them. Then, we show that these models can be easily translated into logic programs, and can be solved by a constraint solver over Boolean variables. In addition, we propose some querying strategies to accelerate the calculation of the grounded, stable and complete extensions. Finally, we describe an experimental study on the performance of our encodings according to different semantics and querying strategies.

We report on the fabrication of a magnetic metallic contaminant detector using multi-channel high-Tc RF-SQUIDs (superconducting quantum interference devices) for large packaged food. For food safety finding small metallic contaminants is an important issue for a food manufacturer. Hence, a detection method for small sized contaminants is required. Some detection systems for food inspection using high-Tc SQUIDs have been reported to date. The system described here is different from the previous systems in its permitted size for inspection, being larger at 150mm in height × 300mm in width. For inspection of large sized food packages, improvement of the signal to noise ratio (SNR) is an important issue because the signal intensity is inversely proportional to the cube of the distance between the SQUID sensor and the object. Therefore a digital filter was introduced and its parameters were optimized. As a result, a steel ball as small as 0.5mm in diameter at a stand-off distance of 167mm was successfully detected with more than SNR = 3.3.

In this paper, a non-isolated bidirectional DC-DC converter with zero voltage switching and constant switching frequency is proposed. Unlike the active clamp bidirectional converters, to create soft switching condition in both direction, only one auxiliary switch is used, which reduces conduction losses and the complexity of the circuit. The proposed converter is controlled by pulse width modulation and the switches are gated complementary, thus the implementation of the control circuit is simple. Low switching losses, high efficiency, high power density, are the advantages of this converter. The simulation and experimental results of the converter verify theoretical analysis. Based on an implemented prototype of the proposed converter at 80 watts, the measured efficiency is 96.5%.

An improved multivariate wavelet denoising algorithm combined with subspace and principal component analysis is presented in this paper. The key element is deriving an optimal orthogonal matrix that can project the multivariate observation signal to a signal subspace from observation space. Univariate wavelet shrinkage operator is then applied to the projected signals channel-wise resulting in the improvement of the output SNR. Finally, principal component analysis is performed on the denoised signal in the observation space to further improve the denoising performance. Experimental results based on synthesized and real world ECG data verify the effectiveness of the proposed algorithm.

Feature extractor plays an important role in visual tracking, but most state-of-the-art methods employ the same feature representation in all scenes. Taking into account the diverseness, a tracker should choose different features according to the videos. In this work, we propose a novel feature adaptive correlation tracker, which decomposes the tracking task into translation and scale estimation. According to the luminance of the target, our approach automatically selects either hierarchical convolutional features or histogram of oriented gradient features in translation for varied scenarios. Furthermore, we employ a discriminative correlation filter to handle scale variations. Extensive experiments are performed on a large-scale benchmark challenging dataset. And the results show that the proposed algorithm outperforms state-of-the-art trackers in accuracy and robustness.

In this work, we study efficient scheduling with network coding in a scalable video coding (SVC) multicast system. Transmission consists of two stages. The original SVC packets are multicasted by the server in the first stage and the lost packets are retransmitted in the second stage. With deadline constraint, the consumer can be only satisfied when the requested packets are received before expiration. Further, the hierarchical encoding architecture of SVC introduces extra decoding delay which poses a challenge for providing acceptable reconstructed video quality. To solve these problems, instantly decodable network coding is applied for reducing the decoding delay, and a novel packet weighted policy is designed to better describe the contribution a packet can make in upgrading the recovered video quality. Finally, an online packet scheduling algorithm based on the maximal weighted clique is proposed to improve the delay, deadline miss ratio and users' experience. Multiple characteristics of SVC packets, such as the packet utility, the slack time and the number of undelivered/wanted packets, are jointly considered. Simulation results prove that the proposed algorithm requires fewer retransmissions and achieves lower deadline miss ratio. Moreover, the algorithm enjoys fine recovery video quality and provides high user satisfaction.

The quality assessment of screen content images (SCIs) has been attractive recently. Different from natural images, SCI is usually a mixture of picture and text. Traditional quality metrics are mainly designed for natural images, which do not fit well into the SCIs. Motivated by this, this letter presents a simple and effective method to naturalize SCIs, so that the traditional quality models can be applied for SCI quality prediction. Specifically, bicubic interpolation-based up-sampling is proposed to achieve this goal. Extensive experiments and comparisons demonstrate the effectiveness of the proposed method.

The Android pattern unlock is a widely adopted graphical password system that requires a user to draw a secret pattern connecting points arranged in a grid. The theoretical security of pattern unlock can be defined by the number of possible patterns. However, only upper bounds of the number of patterns have been known except for 3×3 and 4×4 grids for which the exact number of patterns was found by brute-force enumeration. In this letter, we present the first lower bound by computing the minimum number of visible points from each point in various subgrids.

This paper presents a hybrid push/pull streaming scheme to take advantage of both the interval caching-based push method and the mesh-based pull method. When a new peer joins, a mesh-based pull method is adopted to avoid the overhead to reorganize the structure only if all of its potential preceding peers are likely to leave before the end of its playback. Otherwise, an interval caching-based push method is adopted so that the better performance of the push method can be maintained until it completes the playback. We demonstrate that our proposed scheme outperforms compared with when either the interval caching-based push method or mesh-based pull method is employed alone.

In this review, we present recent advances relating to superconducting nanowire single-photon detectors (SSPDs or SNSPDs) and their broad range of applications. During a period exceeding ten years, the system performance of SSPDs has been drastically improved, and lately excellent detection efficiencies have been realized in practical systems for a wide range of target photon wavelengths. Owing to their advantages such as high system detection efficiency, low dark count rate, and excellent timing jitter, SSPDs have found application in various research fields such as quantum information, quantum optics, optical communication, and also in the life sciences. We summarize the photon detection principle and the current performance status of practical SSPD systems. In addition, we introduce application examples in which SSPDs have been applied.