Cross-Modal Sentiment Sensing with Visual-Augmented Representation and Diverse Decision Fusion.

The rising use of online media has changed the social customs of the public. Users have become accustomed to sharing daily experiences and publishing personal opinions on social networks. Social data carrying emotion and attitude has provided significant decision support for numerous tasks in sentiment analysis. Conventional methods for sentiment classification only concern textual modality and are vulnerable to the multimodal scenario, while common multimodal approaches only focus on the interactive relationship among modalities without considering unique intra-modal information. A hybrid fusion network is proposed in this paper to capture both inter-modal and intra-modal features. Firstly, in the stage of representation fusion, a multi-head visual attention is proposed to extract accurate semantic and sentimental information from textual contents, with the guidance of visual features. Then, multiple base classifiers are trained to learn independent and diverse discriminative information from different modal representations in the stage of decision fusion. The final decision is determined based on fusing the decision supports from base classifiers via a decision fusion method. To improve the generalization of our hybrid fusion network, a similarity loss is employed to inject decision diversity into the whole model. Empiric results on five multimodal datasets have demonstrated that the proposed model achieves higher accuracy and better generalization capacity for multimodal sentiment analysis.

Towards accurate node-based detection of P2P botnets.

Botnets are a serious security threat to the current Internet infrastructure. In this paper, we propose a novel direction for P2P botnet detection called node-based detection. This approach focuses on the network characteristics of individual nodes. Based on our model, we examine node's flows and extract the useful features over a given time period. We have tested our approach on real-life data sets and achieved detection rates of 99-100% and low false positives rates of 0-2%. Comparison with other similar approaches on the same data sets shows that our approach outperforms the existing approaches.

Mitochondrial genome of the neotropical Parnell's mustached bat Pteronotus parnellii (Pteronotus, Mormoopidae).

We reported complete mitochondrial genome of neotropical Parnell's mustached bat Pteronotus parnellii (16,564 bp in length). Its mitochondrial genome consisted of 13 protein-coding genes, 22 transfer RNA genes, 2 ribosomal RNA genes and 1 control region. The base composition of mtDNA was A (33.1%), G (13.7%), C (27.3%), T (25.9%), and the percentage of A and T (59.0%) was higher than G and C. All the genes in P. parnellii were distributed on the H-strand, except for the ND6 subunit gene and 10 tRNA genes, which were encoded on the L-strand. The new sequenced mitochondrial genome will contribute to our understanding of the systematic evolution in the bats.

Measuring wing kinematics, flight trajectory and body attitude during forward flight and turning maneuvers in dragonflies.

A robust technique for determining the wing kinematics, body position and attitude of a free-flight dragonfly is described. The new method is based on a projected comb-fringe technique combined with the natural landmarks on a dragonfly, allowing us to establish the local body-centered coordinate system with high accuracy, and to measure the body attitude at any instant. The kinematic parameters, including wingbeat frequency, flapping angle, angle of attack, torsional angle and camber deformation, required no assumptions to be made with respect to wing geometry, deformability (except the assumption of rigid leading edges) or bilateral wing symmetry. Two typical flight behaviors, forward flight and turning maneuvers, of dragonflies Polycanthagyna melanictera Selys were measured and analyzed.

Step-height measurement by means of a dual-frequency interferometric confocal microscope.

A novel instrument, the dual-frequency interferometric confocal microscope (DICM), which facilitates the measurement of step features, is investigated. It combines the advantages of the high resolution (subnanometer) of heterodyne interferometry and the relatively large measurement range (approximately 5 microm) of confocal microscopy. The axial response curves of the confocal microscopy system are compared in experiments in which microscopic objects with various numerical apertures and magnifications are used. The results prove that the variation in light intensity is enough to permit discrimination of different orders of interference fringes. The DICM has been successfully utilized to measure the step height of a standard mask, and the experimental results agree well with those measured by scanning probe microscopes. The results also show that the system has good repeatability, with a maximum deviation of 5 nm.

Nondestructive measurement of an optical fiber refractive-index profile by a transmitted-light differential interference contact microscope.

A novel transmitted-light differential interference contrast (DIC) system is used for nondestructive measurement of the refractive-index profile (RIP) of an optical fiber. By means of this system the phase of a measured light beam can be modulated with an analyzer, and the phase distribution of a fiber is obtained by calculation of the various interference patterns. The measurement theory and structure and some typical applications of this system are demonstrated. The results of measuring RIPs in graded-index fiber are presented. Both the experimental results and theoretical analysis show that the system takes the advantage of high index resolution and of sufficient measurement accuracy for measuring the refractive index of the optical fiber. The system has strong ability to overcome environmental disturbance because of its common-path design. Moreover, one can use the system to measure the RIP along the fiber axis and acquire an image of the three-dimensional RIP of the fiber.