Understand Students Feedback Using Bi-Integrated CRF Model Based Target Extraction

Educational institutions showing interest to find the opinion of the students about their course and the instructors to enhance the teaching-learning process. For this, most research uses sentiment analysis to track students’ behavior. Traditional sentence-level sentiment analysis focuses on the whole sentence sentiment. Previous studies show that the sentiments alone are not enough to observe the feeling of the students because different words express different sentiments in a sentence. There is a need to extract the targets in a given sentence which helps to find the sentiment towards those targets. Target extraction is the subtask of targeted sentiment analysis. In this paper, we proposed the innovative model to find the targets of the given sentence using Bi-Integrated Conditional Random Fields (CRF). A Parallel fusion neural network model is designed to perform this task. We evaluate the model using the Michigan dataset and we build a dataset for target extraction from student reviews. The experimental results show that our proposed fusion model achieves better results compared to baseline models.     

Classification of brain neoplasm from multi-modality MRI with the aid of ANFIS classifier

Multidimensional Systems and Signal Processing - In medical Image processing, the chief objective is to detect Neoplasm effectively. Neoplasm is basically a sort of abnormal excessive cell growth...     

Rheological and thermal properties of saponified cassava starch‐g‐poly(acrylamide) superabsorbent polymers varying in grafting parameters and absorbency

Cassava starch‐graft‐poly(acrylamide) superabsorbent polymers (SAPs) with varying absorbencies were synthesized. Weight average molecular weight (M_w) of the hydrolyzed starch‐graft‐copolymers ranged from 1.6 × 10⁶ to 2.8 × 10⁶ g/mol, the largest being shown by the sample with highest percentage grafting. The storage (G′) and loss modulus (G″) of hydrogels were determined as a function of frequency. G″ was larger than G′ for the hydrogels with higher absorbencies and exhibited a liquid‐like behavior. However, hydrogels with lower absorbencies showed a reverse viscoelastic behavior. The viscosity of hydrogels determined using a Brookfield viscometer at different shear rates was found to be larger for the hydrogels with higher absorbencies. The melting temperature (T_m) and enthalpy change of fusion (ΔH_f) of the SAPs ranged from 149.7 to 177.7°C and 65 to 494.9 J/g, respectively and showed a positive correlation with grafting parameters and M_w. Heavy metal ion removal capacity of hydrogel followed the order Cu²⁺ > Pb²⁺ > Zn²⁺. © 2014 Wiley Periodicals, Inc. J. Appl. Polym. Sci. 2014 , 131, 40368.     

Multifunctional Dy (III) doped di-calcium silicate array for boosting display and forensic applications

Latent fingerprints (LFPs) are the major physical evidences for the identification of individuals during crime spot investigation. Till date, numerous methods were followed to visualize LFPs. However, simple, accurate, and cost-effective method has wide scope in advanced forensic field. In our work, Ca₂SiO₄:Dy³⁺ nanopowders (NPs) were fabricated via solution combustion route. The optimized sample was employed for the visualization of overlapped LFPs by the cost effective powder dusting method. The obtained results reveals the complete three levels of ridge characteristics with high sensitivity, reproducibility, selectivity, and reliability on various complex surfaces. The photoluminescence (PL) spectra consist of intense peaks at ～ 480 and 574 nm owing to ⁴F_9/2 → ⁶H_15/2 and ⁴F_9/2 → ⁶H_13/2 4f transitions of Dy³⁺ ions, respectively. The photometric properties confirm that the samples exhibit intense white emission with high color purity. Therefore, the prepared NPs could be a definitive choice as an advanced luminescent NPs for forensic, solid state lighting and portable FED devices.     

Prediction of the tensile capacity of cold formed angles experiencing shear lag

The present study deals with the shear lag phenomenon in cold formed angles under tension, which are connected on one leg. A new expression for shear lag factor which represents the net section reduction coefficient has been suggested in the present paper. The proposed expression based on the regression analysis of 108 experimental results reported in the literature is validated by experiments involving net section failure in angles under tension. Totally 18 experiments were carried out on single angles fastened with bolts to the gusset plates under tension loaded upto net section rupture mode of failure. The experimental test parameters considered are number of bolts, pitch and shear lag distances and ratio of connected leg length to unconnected leg length. The tensile capacities are evaluated by various specifications such as AS/NZS:4600:2005, NAS:2001, AISC:2005, BS:5950-Part5:1998, IS:800-2007 and the proposed equation. A comparative study of tensile capacities predicted based on various codes and the experiment results is presented in this paper. For the tested range of specimens, both NAS:2001 and AISC:2005 standards over-predicted the capacities for all the specimens. The IS:800-2007 and AS/NZS 4600:2005 predictions are good for the specimens with three bolted connections and unconservative in the case of specimen with two bolts. Both BS:5950-Part-5:1998 and the proposed equation for IS:801 predict good estimate of the tensile capacity of cold formed angle members. The proposed equation for cold formed steel tension members, which is in the same format of IS:800 (2007) (Indian code for Hot rolled steel design), has been demonstrated to be good.     

Advances In Peer-To-Peer Content Search

Peer-to-peer (P2P) computer networks have recently received tremendous attention due to their inherent scalability and flexibility, which facilitates a broad spectrum of innovative multimedia applications. Such networks rely on the power of participant nodes of the network (called peers) for communications and computation. Traditional applications of P2P multimedia include decentralized file sharing and content distribution. Yet, the value of the virtually unlimited amount of data distributed in the P2P network will be sacrificed if effective and efficient ways to locate the content are missing. This challenge has stimulated extensive research in recent years, and many new P2P content search methods have been proposed. This paper provides a timely review of influential work in the area of peer-to-peer (P2P) content search. We begin with a survey of text-based P2P search mechanisms and continue with an exposition of content-based and semantic-based approaches followed by a discussion of future directions.     

A comprehensive computer program package for small signal stabilityanalysis of power systems

A package of integrated programs for small-signal stability analysis of large interconnected power systems is described. The package has extensive modeling capability and uses alternative eigenvalue calculation techniques, making it suitable for the analysis of a wide range of stability and control problems. Results of eigenvalue calculations for three power systems of differing size and complexity are presented and the accuracy, consistency and convergence of the alternative calculation methods are discussed     

Analysis and design of watermarking algorithms for improved resistance to compression

We study the performance of robust digital watermarking approaches in the presence of lossy compression by introducing practical analysis methodologies. Correlation expressions between the embedded watermark and the extracted watermark are derived to determine the optimal watermarking domain to maximize data hiding rates for spread spectrum and quantization watermarking. It is determined both theoretically and through simulations that the embedding strategy, in addition to the transform used for lossy compression, dictate the optimal transform for watermarking. Through analytic comparisons, we develop a novel hybrid watermarking algorithm that exploits the best of both approaches for greater resilience to JPEG compression.     

Spatially localized image-dependent watermarking for statistical invisibility and collusion resistance

We develop a novel video watermarking framework based on the collusion-resistant design rules formulated in a companion paper. We propose to employ a spatially-localized image dependent approach to create a watermark whose pairwise frame correlations approximate those of the host video. To characterize the spread of its spatially-localized energy distribution, the notion of a watermark footprint is introduced. Then we explain how a particular type of image dependent footprint structure, comprised of subframes centered around a set of visually significant anchor points, can lead to two advantageous results: pairwise watermark frame correlations that more closely match those of the host video for statistical invisibility, and the ability to apply image watermarks directly to a frame sequence without sacrificing collusion-resistance. In the ensuing overview of the proposed video watermark, two new ideas are put forward: synchronizing the subframe locations using visual content rather than structural markers and exploiting the inherent spatial diversity of the subframe-based watermark to improve detector performance. Simulation results are presented to show that the proposed scheme provides improved resistance to linear frame collusion, while still being embedded and extracted using relatively low complexity frame-based algorithms.