In practice, emotional speech utterances are often collected from different devices or conditions, which will lead to discrepancy between the training and testing data, resulting in sharp decrease of recognition rates. To solve this problem, in this letter, a novel transfer semi-supervised non-negative matrix factorization (TSNMF) method is presented. A semi-supervised negative matrix factorization algorithm, utilizing both labeled source and unlabeled target data, is adopted to learn common feature representations. Meanwhile, the maximum mean discrepancy (MMD) as a similarity measurement is employed to reduce the distance between the feature distributions of two databases. Finally, the TSNMF algorithm, which optimizes the SNMF and MMD functions together, is proposed to obtain robust feature representations across databases. Extensive experiments demonstrate that in comparison to the state-of-the-art approaches, our proposed method can significantly improve the cross-corpus recognition rates.

A vehicular ad hoc network (VANET) consists of vehicles (mobile nodes) and road side units which are equipped with the wireless devices such as wireless LANs. Mobile nodes exchange information messages with each other so that VANETs are configured in a self-organized manner. As one of network service scenarios in VANETs, there is a network service to provide the parking spaces in the city central to vehicles (mobile nodes). In this scenario, the road side units (source nodes) which are deployed at the parking spaces periodically disseminate the number of available parking spaces to mobile nodes. Therefore, in this paper, we propose a mobile agent-based information dissemination scheme using location information of mobile nodes and source nodes in the VANET environment. In addition, we conduct simulation experiments in the VANET environment to evaluate the proposed mobile agent-based information dissemination scheme. We confirmed that it could disseminate information messages with lower overhead because mobile agents migrate among mobile nodes by using the location information.

The automation of traceability links or traceability matrices is important to many software development paradigms. In turn, the efficiency and effectiveness of the recovery of traceability links in the distributed software development is becoming increasingly vital due to complexity of project developments, as this include continuous change in requirements, geographically dispersed project teams, and the complexity of managing the elements of a project - time, money, scope, and people. Therefore, the traceability links among the requirements artifacts, which fulfill business objectives, is also critical to reduce the risk and ensures project‘s success. This paper proposes Autonomous Decentralized Semantic based Traceability Link Recovery (AD-STLR) architecture. According to best of our knowledge this is the first architectural approach that uses an autonomous decentralized concept, DBpedia knowledge-base, Babelnet 2.5 multilingual dictionary and semantic network, for finding similarity among different project artifacts and the automation of traceability links recovery.

In the current IoT (Internet of Things) environment, more and more Things: devices, objects, sensors, and everyday items not usually considered computers, are connected to the Internet, and these Things affect and change our social life and economic activities. By using IoTs, service providers can collect and store personal information in the real world, and such providers can gain access to detailed behaviors of the user. Although service providers offer users new services and numerous benefits using their detailed information, most users have concerns about the privacy and security of their personal data. Thus, service providers need to take countermeasures to eliminate those concerns. To help eliminate those concerns, first we conduct a survey regarding users' privacy and security concerns about IoT services, and then we analyze data collected from the survey using structural equation modeling (SEM). Analysis of the results provide answers to issues of privacy and security concerns to service providers and their users. And we also analyze the effectiveness and effects of personal information management and protection functions in IoT services.

Mapping instances to the Linked Open Data (LOD) cloud plays an important role for enriching information of instances, since the LOD cloud contains abundant amounts of interlinked instances describing the instances. Consequently, many techniques have been introduced for mapping instances to a LOD data set; however, most of them merely focus on tackling with the problem of heterogeneity. Unfortunately, the problem of the large number of LOD data sets has yet to be addressed. Owing to the number of LOD data sets, mapping an instance to a LOD data set is not sufficient because an identical instance might not exist in that data set. In this article, we therefore introduce a heuristic expansion based framework for mapping instances to LOD data sets. The key idea of the framework is to gradually expand the search space from one data set to another data set in order to discover identical instances. In experiments, the framework could successfully map instances to the LOD data sets by increasing the coverage to 90.36%. Experimental results also indicate that the heuristic function in the framework could efficiently limit the expansion space to a reasonable space. Based upon the limited expansion space, the framework could effectively reduce the number of candidate pairs to 9.73% of the baseline without affecting any performances.

Sentence similarity computation is an increasingly important task in applications of natural language processing such as information retrieval, machine translation, text summarization and so on. From the viewpoint of information theory, the essential attribute of natural language is that the carrier of information and the capacity of information can be measured by information content which is already successfully used for word similarity computation in simple ways. Existing sentence similarity methods don't emphasize the information contained by the sentence, and the complicated models they employ often need using empirical parameters or training parameters. This paper presents a fully unsupervised computational model of sentence semantic similarity. It is also a simply and straightforward model that neither needs any empirical parameter nor rely on other NLP tools. The method can obtain state-of-the-art experimental results which show that sentence similarity evaluated by the model is closer to human judgment than multiple competing baselines. The paper also tests the proposed model on the influence of external corpus, the performance of various sizes of the semantic net, and the relationship between efficiency and accuracy.

The growing trends in Internet usage for data and knowledge sharing calls for dynamic classification of web contents, particularly at the edges of the Internet. Rather than considering Linked Data as an integral part of Big Data, we propose Autonomous Decentralized Semantic-based Content Classifier (ADSCC) for dynamic classification of unstructured web contents, using Linked Data and web metadata in Content Delivery Network (CDN). The proposed framework ensures efficient categorization of URLs (even overlapping categories) by dynamically mapping the changing user-defined categories to ontologies' category/classes. This dynamic classification is performed by the proposed system that mainly involves three main algorithms/modules: Dynamic Mapping algorithm, Autonomous coordination-based Inference algorithm, and Context-based disambiguation. Evaluation results show that the proposed system achieves (on average), the precision, recall and F-measure within the 93-97% range.

In this study, our recent research activities on nanophotonic devices with semiconductor quantum nanostructures are reviewed. We have developed a technique for nanofabricating of high-quality and high-density semiconductor quantum dots (QDs). On the basis of this core technology, we have studied next-generation nanophotonic devices fabricated using high-quality QDs, including (1) a high-performance QD laser for long-wavelength optical communications, (2) high-efficiency compound-type solar cell structures, and (3) single-QD devices for future applications related to quantum information. These devices are expected to be used in high-speed optical communication systems, high-performance renewable energy systems, and future high-security quantum computation and communication systems.

This paper presents novel reconfigurable semi-systolic array architecture for the Smith-Waterman with an affine gap penalty algorithm to align protein sequences optimized for shorter database sequences. This architecture has been modified to enable hardware reuse rather than replicating processing elements of the semi-systolic array in multiple FPGAs. The proposed hardware architecture and the previously published conventional one are described at the Register Transfer Level (RTL) using VHDL language and implemented using the FPGA technology. The results show that the proposed design has significant higher normalized speedup (up to 125%) over the conventional one for query sequence lengths less than 512 residues. According to the UniProtKB/TrEMBL protein database (release 2015_05) statistics, the largest number of sequences (about 80%) have sequence length less than 512 residues that makes the proposed design outperforms the conventional one in terms of speed and area in this sequence lengths range.

We developed an electrically driven near-infrared broadband light source based on self-assembled InAs quantum dots (QDs). By combining emissions from four InAs QD ensembles with controlled emission center wavelengths, electro-luminescence (EL) with a Gaussian-like spectral shape and approximately 85-nm bandwidth was obtained. The peak wavelength of the EL was blue-shifted from approximately 1230 to 1200 nm with increased injection current density (J). This was due to the state-filling effect: sequential filling of the discrete QD electron/hole states by supplied carriers from lower (ground state; GS) to higher (excited state; ES) energy states. The EL intensities of the ES and GS emissions exhibited different J dependence, also because of the state-filling effect. The point-spread function (PSF) deduced from the Fourier-transformed EL spectrum exhibited a peak without apparent side lobes. The half width at half maximum of the PSF was 6.5 µm, which corresponds to the estimated axial resolution of the optical coherence tomography (OCT) image obtained with this light source. These results demonstrate the effectiveness of the QD-based device for realizing noise-reduced high-resolution OCT.

Disassembly, as a principal reverse-engineering tool, is the process of recovering the equivalent assembly instructions of a program's machine code from its binary representation. However, when disassembling a firmware file, the disassembly process cannot be performed well if the image base is unknown. In this paper, we propose an innovative method to determine the image base of a firmware file with ARM/Thumb instruction set. First, based on the characteristics of the function entry table (FET) for an ARM processor, an algorithm called FIND-FET is proposed to identify the function entry tables. Second, by using the most common instructions of function prologue and FETs, the FIND-BASE algorithm is proposed to determine the candidate image base by counting the matched functions and then choose the one with maximal matched FETs as the final result. The algorithms are applied on some firmwares collected from the Internet, and results indicate that they can effectively find out the image base for the majority of example firmware files.

Recent small cube satellites use higher frequency bands such as Ku-band for higher throughput communications. This requires high-frequency link in an earth radio station as well. As one of the solutions, we propose usage of bidirectional radio-on-fiber link employing a wavelength multiplexing scheme. It was numerically shown that the response linearity of the electro-absorption modulator integrated laser (EML) is sufficient and that the spurious emissions are lower enough or can be reduced by the radio-frequency filters. From the frequency response and the single-sideband phase noise measurements, the EML was proved to be used in a radio-on-fiber system of the cube satellite earth station.

We review recent progress in integrated photonics devices and their applications for datacom. In addition to current technology used in 100-Gigabit Ethernet (100GbE) with a compact form-factor of the transceiver, the next generation of technology for 400GbE seeks a larger number of wavelengths with a more sophisticated modulation format and higher bit rate per wavelength. For wavelength scalability and functionality, planar lightwave circuits (PLCs), such as arrayed waveguide gratings (AWGs), will be important, as well higher-order-modulation to ramp up the total bit rate per wavelength. We introduce integration technology for a 100GbE optical sub-assembly that has a 4λ x 25-Gb/s non-return-to-zero (NRZ) modulation format. For beyond 100GbE, we also discuss applications of 100GbE sub-assemblies that provide 400-Gb/s throughput with 16λ x 25-Gb/s NRZ and bidirectional 8λ x 50-Gb/s four-level pulse amplitude modulation (PAM4) using PLC cyclic AWGs.

Boneh and Franklin considered to add the revocation functionality to identity-based encryption (IBE). Though this methodology is applicable to any IBE and hierarchical IBE (HIBE), the resulting scheme is non-scalable. Therefore, a generic transformation of scalable revocable (H)IBE (R(H)IBE) from non-scalable R(H)IBE is really desirable. Towards this final goal, in this paper we introduce prototype RHIBE which does not require to be scalable (but requires some conditions), and propose a generic transformation of scalable RHIBE from prototype RHIBE. Moreover, we construct a prototype RHIBE scheme based on the decisional bilinear Diffie-Hellman (DBDH) assumption. Since our prototype RHIBE provides history-free update, insider security, and decryption key exposure resistance, our construction yields the first RHIBE scheme based on the static assumption with these desirable properties.

In many movies, social conditions and awareness of the issues of the times are depicted in any form. Even if fantasy and science fiction are works far from reality, the character relationship does mirror the real world. Therefore, we try to understand social conditions of the real world by analyzing the movie. As a way to analyze the movies, we propose a method of estimating interpersonal relationships of the characters, using a machine learning technique called Markov Logic Network (MLN) from movie script databases on the Web. The MLN is a probabilistic logic network that can describe the relationships between characters, which are not necessarily satisfied on every line. In experiments, we confirmed that our proposed method can estimate favors between the characters in a movie with F-measure of 58.7%. Finally, by comparing the relationships with social indicators, we discussed the relevance of the movies to the real world.

Recent work on joint word segmentation, POS (Part Of Speech) tagging, and dependency parsing in Chinese has two key problems: the first is that word segmentation based on character and dependency parsing based on word were not combined well in the transition-based framework, and the second is that the joint model suffers from the insufficiency of annotated corpus. In order to resolve the first problem, we propose to transform the traditional word-based dependency tree into character-based dependency tree by using the internal structure of words and then propose a novel character-level joint model for the three tasks. In order to resolve the second problem, we propose a novel semi-supervised joint model for exploiting n-gram feature and dependency subtree feature from partially-annotated corpus. Experimental results on the Chinese Treebank show that our joint model achieved 98.31%, 94.84% and 81.71% for Chinese word segmentation, POS tagging, and dependency parsing, respectively. Our model outperforms the pipeline model of the three tasks by 0.92%, 1.77% and 3.95%, respectively. Particularly, the F1 value of word segmentation and POS tagging achieved the best result compared with those reported until now.

The objective of this paper is to recognize and classify the poses of idols in still images on the web. The poses found in Japanese idol photos are often complicated and their classification is highly challenging. Although advances in computer vision research have made huge contributions to image recognition, it is not enough to estimate human poses accurately. We thus propose a method that refines result of human pose estimation by Pose Guide Ontology (PGO) and a set of energy functions. PGO, which we introduce in this paper, contains useful background knowledge, such as semantic hierarchies and constraints related to the positional relationship between body parts. Energy functions compute the right positions of body parts based on knowledge of the human body. Through experiments, we also refine PGO iteratively for further improvement of classification accuracy. We demonstrate pose classification into 8 classes on a dataset containing 400 idol images on the web. Result of experiments shows the efficiency of PGO and the energy functions; the F-measure of classification is 15% higher than the non-refined results. In addition to this, we confirm the validity of the energy functions.

We investigated a control of the crystalline orientation of soluble organic semiconductor single crystals using liquid crystal solvents aligned by the electric field to improve the performance of organic thin-film transistors. We clarified that the semiconductor single crystal grows to the direction parallel to the liquid crystal alignment oriented by the lateral electric field.

When there are multiple component predictors, it is promising to integrate them into one predictor for advanced reasoning. If each component predictor is given as a stochastic model in the form of probability distribution, an exponential mixture of the component probability distributions provides a good way to integrate them. However, weight parameters used in the exponential mixture model are difficult to estimate if there is no training samples for performance evaluation. As a suboptimal way to solve this problem, weight parameters may be estimated so that the exponential mixture model should be a balance point that is defined as an equilibrium point with respect to the distance from/to all component probability distributions. In this paper, we propose a weight parameter estimation method that represents this concept using a symmetric Kullback-Leibler divergence and generalize this method.

For the last few decades, learning with multiple kernels, represented by the ensemble kernel regressor and the multiple kernel regressor, has attracted much attention in the field of kernel-based machine learning. Although their efficacy was investigated numerically in many works, their theoretical ground is not investigated sufficiently, since we do not have a theoretical framework to evaluate them. In this paper, we introduce a unified framework for evaluating kernel regressors with multiple kernels. On the basis of the framework, we analyze the generalization errors of the ensemble kernel regressor and the multiple kernel regressor, and give a sufficient condition for the ensemble kernel regressor to outperform the multiple kernel regressor in terms of the generalization error in noise-free case. We also show that each kernel regressor can be better than the other without the sufficient condition by giving examples, which supports the importance of the sufficient condition.