Toward an efficient and scalable feature selection approach for internet traffic classification

There is significant interest in the network management and industrial security community about the need to identify the “best” and most relevant features for network traffic in order to properly characterize user behaviour and predict future traffic. The ability to eliminate redundant features is an important Machine Learning (ML) task because it helps to identify the best features in order to improve the classification accuracy as well as to reduce the computational complexity related to the construction of the classifier. In practice, feature selection (FS) techniques can be used as a preprocessing step to eliminate irrelevant features and as a knowledge discovery tool to reveal the “best” features in many soft computing applications. In this paper, we investigate the advantages and disadvantages of such FS techniques with new proposed metrics (namely goodness, stability and similarity). We continue our efforts toward developing an integrated FS technique that is built on the key strengths of existing FS techniques. A novel way is proposed to identify efficiently and accurately the “best” features by first combining the results of some well-known FS techniques to find consistent features, and then use the proposed concept of support to select a smallest set of features and cover data optimality. The empirical study over ten high-dimensional network traffic data sets demonstrates significant gain in accuracy and improved run-time performance of a classifier compared to individual results produced by some well-known FS techniques.     

Supporting multiple perspectives in feature-based configuration

Feature diagrams have become commonplace in software product line engineering as a means to document variability early in the life cycle. Over the years, their application has also been extended to assist stakeholders in the configuration of software products. However, existing feature-based configuration techniques offer little support for tailoring configuration views to the profiles of the various stakeholders. In this paper, we propose a lightweight, yet formal and flexible, mechanism to leverage multidimensional separation of concerns in feature-based configuration. We propose a technique to specify concerns in feature diagrams and to generate automatically concern-specific configuration views. Three alternative visualisations are proposed. Our contributions are motivated and illustrated through excerpts from a real web-based meeting management application which was also used for a preliminary evaluation. We also report on the progress made in the development of a tool supporting multi-view feature-based configuration.     

圆锯片综合特性检测系统的硬件设计

圆锯片综合特性检测系统是根据国内外学者的圆锯片适张研究成果和频谱分析、坐标映射的理论,综合计算机技术、传感器技术、信号调理技术和数据采集处理技术来开发的,检测系统软件是Visual C 开发,在Win32环境下运行.整个检测系统操作简便、人机界面友好、具有较强数据管理功能,能在工业现场条件下快速、准确、无损地检测圆锯片适张状态.     

The biological properties of Aerococcus antagonists, representatives of human microbiocenoses

The paper deals with the data on biology of Aerococcus, a slightly studied group of microorganisms. Physiological-biochemical properties of Aerococcus are described, data of their distribution in nature are given. Peculiar attention is paid to the estimate of the role of Aerococcus in human microbiocenoses. As a result of the profound and all-round study of this group of microorganisms the authors have developed new bacterial drug "A-bakterin" based on the aerococcus strain. Data presented about the results of clinical tests of "A-bakterin" are presented, a possibility to use Aerococcus lysate in the elaboration of new drugs is discussed.     

Recent advances in amperometric glucose biosensors for in vivo monitoring

The substance-use, positive health practices, diet and eating habits, driving behaviours and preventive health care practices of 113 nurses living in the southeast of England were assessed by means of a postal questionnaire. Females were generally more compliant than males with most of the behaviours assessed. Females reported taking greater risks with driving behaviours by an occasional tendency to "drink and drive". Many males reported both a lack of knowledge of how to self-examine their testicles for signs of cancer and little compliance with this behaviour. This sample reported more smoking and alcohol use than previously reported samples of American nurses and U.K. norms. The nurses in this sample were more likely to avoid foods high in cholesterol, add fibre to their diet and be more vigilant in their driving behaviour than American nurses and British non-nurses.     

Clinical and epidemiological findings during a measles outbreak occurring in a population with a high vaccination coverage

Malignant mesothelioma is caused almost exclusively by occupational exposure to asbestos. During the past few years, however, increasing evidence has mounted that background exposure to asbestos could be sufficient to cause mesothelioma. Treatment of malignant mesothelioma remains a big problem. Some new approaches are on their way, and the most exciting ones are local immunotherapy in very early cases. Some success has been reported with local interferon treatment. As for treatment of metastatic pleural disease, the main purpose is symptomatic relief of dyspnea caused by fluid accumulation. The best way to achieve a lasting palliation is pleurodesis, and the most common way to do this, is by chemical means. The drug of choice in the United States has for many years been tetracycline, but since injectable tetracycline is no longer available, some substitute must be found. The substance that will "win" is not yet clear, but the two leading contestants are talc and doxycycline. Bleomycin also has its supporters, and a dark horse is quinacrine, which although not easily available in the United States, has been used in many European centers for decades.     

Is Deep-Learning and Natural Language Processing Transcending the Financial Forecasting? Investigation Through Lens of News Analytic Process

Computational Economics - This study tries to unravel the stock market prediction puzzle using the textual analytic with the help of natural language processing (NLP) techniques and Deep-learning...     

Somatic mosaicism in a patient with neurofibromatosis type 1

Using loss of heterozygosity analysis, a method designed to detect moderate to large gene deletions, we have identified a new-mutation neurofibromatosis type 1 (NF1) patient who is somatically mosaic for a large maternally derived deletion in the NF1 gene region. The deletion extends at least from exon 4 near the 5' end of the gene to intron 39 near the 3' end. The gene-coding region is, therefore, mostly or entirely deleted, encompassing a loss of > or = 100 kb. We hypothesize that the deletion occurred at a relatively early developmental timepoint, since signs of NF1 in this patient are not confined to a specific body region, as seen in "segmental" NF, and since both mesodermally and ectodermally derived cells are affected. This report provides the first molecular evidence of somatic mosaicism in NF1 and, taken together with a recent report of germ-line mosaicism in NF1, adds credence to the concept that mosaicism plays an important role in phenotypic and genetic aspects of NF1 and may even be a relatively common phenomenon.     

A novel composite model approach to improve software quality prediction

Context:How can quality of software systems be predicted before deployment? In attempting to answer this question, prediction models are advocated in several studies. The performance of such models drops dramatically, with very low accuracy, when they are used in new software development environments or in new circumstances.ObjectiveThe main objective of this work is to circumvent the model generalizability problem. We propose a new approach that substitutes traditional ways of building prediction models which use historical data and machine learning techniques.MethodIn this paper, existing models are decision trees built to predict module fault-proneness within the NASA Critical Mission Software. A genetic algorithm is developed to combine and adapt expertise extracted from existing models in order to derive a “composite” model that performs accurately in a given context of software development. Experimental evaluation of the approach is carried out in three different software development circumstances.ResultsThe results show that derived prediction models work more accurately not only for a particular state of a software organization but also for evolving and modified ones.ConclusionOur approach is considered suitable for software data nature and at the same time superior to model selection and data combination approaches. It is then concluded that learning from existing software models (i.e., software expertise) has two immediate advantages; circumventing model generalizability and alleviating the lack of data in software-engineering.