Semi-supervised classification based on subspace sparse representation

Graph plays an important role in graph-based semi-supervised classification. However, due to noisy and redundant features in high-dimensional data, it is not a trivial job to construct a well-structured graph on high-dimensional samples. In this paper, we take advantage of sparse representation in random subspaces for graph construction and propose a method called Semi-Supervised Classification based on Subspace Sparse Representation, SSC-SSR in short. SSC-SSR first generates several random subspaces from the original space and then seeks sparse representation coefficients in these subspaces. Next, it trains semi-supervised linear classifiers on graphs that are constructed by these coefficients. Finally, it combines these classifiers into an ensemble classifier by minimizing a linear regression problem. Unlike traditional graph-based semi-supervised classification methods, the graphs of SSC-SSR are data-driven instead of man-made in advance. Empirical study on face images classification tasks demonstrates that SSC-SSR not only has superior recognition performance with respect to competitive methods, but also has wide ranges of effective input parameters.

Integrated RNA extraction and RT-PCR for semi-quantitative gene expression studies on a microfluidic device

This paper describes the development of a microfluidic methodology, using RNA extraction and reverse transcription PCR, for investigating expression levels of cytochrome P450 genes. Cytochrome P450 enzymes are involved in the metabolism of xenobiotics, including many commonly prescribed drugs, therefore information on their expression is useful in both pharmaceutical and clinical settings. RNA extraction, from rat liver tissue or primary rat hepatocytes, was performed using a silica-based solid-phase extraction technique. Following elution of the purified RNA, amplification of target sequences for the housekeeping gene, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) and the cytochrome P450 gene CYP1A2, was carried out using a one-step reverse transcription PCR. Once the microfluidic methodology had been optimized, analysis of control and 3-methylcholanthrene-induced primary rat hepatocytes were used to evaluate the system. As expected, GAPDH was consistently expressed, whereas CYP1A2 levels were found to be raised in the drug-treated samples. The proposed system offers an initial platform for development of both rapid throughput analyzers for pharmaceutical drug screening and point-of-care diagnostic tests to aid provision of drug regimens, which can be tailor-made to the individual patient.

Performance study of a wind-grid system in a semi- arid region and major integration issues

The phenomenal growth in economy experienced in developed countries throughout the 20th century has largely been driven by the availability of conventional energy sources for electricity generation. However, increased concern about fossil fuels and adverse effect of carbon dioxide emission in to atmosphere changed the conventional power system to a viable one by integrating renewable energy sources into the existing system. Among the Renewable Energy (RE) sources, wind energy is one of the fastest growing technologies in reducing the Green House Gas (GHG) emissions in to the atmosphere due to its continuous availability throughout a period. Hence, this paper discusses the performance of a wind-grid connected system in a semi-arid region by conducting a case study. Wilson promontory, one of the best locations for wind generation in Victoria is considered as a case study. Hybrid Optimization Model for Electric Renewable (HOMER) is used as a simulating tool for this analysis. This study also presents the influences of storage system in the proposed Hybrid Power System (HPS) allowing energy to be stored during higher generations or lower load demands. In addition this paper also discusses the major integration issues to facilitate the large scale wind energy into the grid for reliable power generation and distribution.

Lithium storage in disordered graphitic materials: a semi-quantitative study of the relationship between structure disordering and capacity.

The application of the graphitic anode is restricted by its low theoretical specific capacity of 372 mA h g(-1). Higher capacity can be achieved in the graphitic anode by modifying its structure, but the detailed storage mechanism is still not clear. In this work, the mechanism of the lithium storage in a disordered graphitic structure has been systematically studied. It is found that the enhanced capacity of the distorted graphitic structure does not come from lithium-intercalation, but through a capacitive process, which depends on the disordering degree and the porous structure.

Prospects of renewable energy in semi-arid region

Continuous usage of fossil fuels and other conventional resources to meet the growing demand has resulted in in-creased energy crisis and greenhouse gas emissions. Hence, it is essential to use renewable energy sources for more reliable, effective, sustainable and pollution free transmission and distribution networks. Therefore, to facilitate large-scale integration of renewable energy in particular wind and solar photovoltaic (PV) energy, this paper presents the feasibility analysis for semi-arid climate and finds the most suitable places in North East region of Victoria for re-newable energy generation. For economic and environmental analysis, Hybrid Optimization Model for Electric Re-newables (HOMER) has used to investigate the prospects of wind and solar energy considering the Net Present Cost (NPC), Cost of Energy (COE) and Renewable fraction (RF). Six locations are selected from North East region of Victo-ria and simulations are performed. From the feasibility analysis, it can be concluded that Mount Hotham is one of the most suitable locations for wind energy generation while Wangaratta is the most suitable location for solar energy generation. Mount Hotham is also the best suitable locations in North East region for hybrid power systems i.e., com-bination of both wind and solar energy generation.

Mixing vector construction for single channel semi-blind source separation using empirical mode decomposition

The Empirical Mode Decomposition (EMD) method is a commonly used method for solving the problem of single channel blind source separation (SCBSS) in signal processing. However, the mixing vector of SCBSS, which is the base of the EMD method, has not yet been effectively constructed. The mixing vector reflects the weights of original signal sources that form the single channel blind signal source. In this paper, we propose a novel method to construct a mixing vector for a single channel blind signal source to approximate the actual mixing vector in terms of keeping the same ratios between signal weights. The constructed mixing vector can be used to improve signal separations. Our method incorporates the adaptive filter, least square method, EMD method and signal source samples to construct the mixing vector. Experimental tests using audio signal evaluations were conducted and the results indicated that our method can improve the similar values of sources energy ratio from 0.2644 to 0.8366. This kind of recognition is very important in weak signal detection.

Propagation model of smartphone worms based on semi-Markov process and social relationship graph

Smartphone applications are getting more and more popular and pervasive in our daily life, and are also attractive to malware writers due to their limited computing source and vulnerabilities. At the same time, we possess limited understanding of our opponents in cyberspace. In this paper, we investigate the propagation model of SMS/MMS-based worms through integrating semi-Markov process and social relationship graph. In our modeling, we use semi-Markov process to characterize state transition among mobile nodes, and hire social network theory, a missing element in many previous works, to enhance the proposed mobile malware propagation model. In order to evaluate the proposed models, we have developed a specific software, and collected a large scale real-world data for this purpose. The extensive experiments indicate that the proposed models and algorithms are effective and practical. © 2014 Elsevier Ltd. All rights reserved.

Towards designing an email classification system using multi-view based semi-supervised learning

The goal of email classification is to classify user emails into spam and legitimate ones. Many supervised learning algorithms have been invented in this domain to accomplish the task, and these algorithms require a large number of labeled training data. However, data labeling is a labor intensive task and requires in-depth domain knowledge. Thus, only a very small proportion of the data can be labeled in practice. This bottleneck greatly degrades the effectiveness of supervised email classification systems. In order to address this problem, in this work, we first identify some critical issues regarding supervised machine learning-based email classification. Then we propose an effective classification model based on multi-view disagreement-based semi-supervised learning. The motivation behind the attempt of using multi-view and semi-supervised learning is that multi-view can provide richer information for classification, which is often ignored by literature, and semi-supervised learning supplies with the capability of coping with labeled and unlabeled data. In the evaluation, we demonstrate that the multi-view data can improve the email classification than using a single view data, and that the proposed model working with our algorithm can achieve better performance as compared to the existing similar algorithms.