Sentiment classification: a lexical similarity based approach for extracting subjectivity in documents

With the growth of social media, document sentiment classification has become an active area of research in this decade. It can be viewed as a special case of topical classification applied only to subjective portions of a document (sources of sentiment). Hence, the key task in document sentiment classification is extracting subjectivity. Existing approaches to extract subjectivity rely heavily on linguistic resources such as sentiment lexicons and complex supervised patterns based on part-of-speech (POS) information. This makes the task of subjective feature extraction complex and resource dependent. In this work, we try to minimize the dependency on linguistic resources in sentiment classification. We propose a simple and statistical methodology called review summary (RSUMM) and use it in combination with well-known feature selection methods to extract subjectivity. Our experimental results on a movie review dataset prove the effectiveness of the proposed methodology.     

Neutrinos and our Sun — Part 1

The elusive neutrinos have periodically yielded their secrets to man and at each such juncture major advances have been achieved in our understanding of the sub-atomic phenomena. These particles also carry invaluable information about the centre of the Sun where energy is generated through nuclear fusion. In Part I of this article, history of the discovery of neutrinos is traced, their properties and types are described. Also, the Standard Model which forms the basis of the structure of matter and of which massless neutrinos are an integral part, is described. The role of neutrinos in solar energy generation and the great ‘solar neutrino puzzle’ and its solution will be described later in the series.     

Influence of medium components on elastase production using crude sources by<Emphasis Type="Italic">Bacillus</Emphasis> sp. EL31410

A newly isolated strain EL31410, producing elastase (E. C3. 4. 4. 7) with high elastolytic activity was identified asBacillus sp. In the medium optimization, it was found that wheat bran and soybean flour hydrosate were the best crude carbon and nitrogen source for enzyme production, respectively. Addition of corn steep flour can affect the bacterium growth and elastase production. A fractional factorial design was applied to study the main factors that affect the enzyme production, and central composite experimental design and response surface methodology were adopted to derive a statistical model for the effect of wheat bran and soybean flour hydrosate on elastase production. The experimental results showed that wheat bran had positive effect but soybean flour hydrosate had negative effect, on enzyme production. An initial concentration of 3.4% (w/v) wheat bran and 9.4% (v/v) soybean flour hydrosate were found to be optimal for enzyme production in batch culture. The time course of elastase production in the optimized medium composition was also described. Project (No. 300024) supported by the Zhejiang Provincial Natural Science Foundation of China     

Spyware – the ethics of covert software

Many computer users are happy to be oblivious of the workings within the machine and yet on some level it is important to know what is occurring therein. This paper discusses an unusual type of surveillance software which may be installed in many computers. The strange aspect of this software is that it has often been downloaded and installed by the user, but without her knowledge. The software is mainly designed to collect information about the user of the computer and relay this information back to the software manufacturer. The download, installation, data collection and data transfer all take place within the users own computer but very seldom with the users knowledge. It is the intention of this paper to describe the technology involved and thereafter discuss how this new technology is affecting the online privacy debate. The paper continues to discuss the basis for legitimacy of technology and also the current level of technological deterrents available. The paper concludes with a comparison of two approaches at resolving the current problem, via legislation or the market approach.     

Die Beweisentwicklungsumgebung W\Omega -Mkrp

Zusammenfassung. Die Beweisentwicklungsumgebung -Mkrpsoll Mathematiker bei einer ihrer Hauptt?tigkeiten, n?mlich dem Beweisen mathematischer Theoreme unterstützen. Diese Unterstützung mu? so komfortabel sein, da? die rechnergestützte Suche nach formalen Beweisen leichter und insbesondere weniger aufwendig ist, als ohne das System. Dazu mu? die verwendete Objektsprache ausdrucksstark sein, man mu? die M?glichkeit haben, abstrakt über Beweispl?ne zu reden, die gefundenen Beweise müssen in einer am Menschen orientierte Form pr?sentiert werden und vor allem mu? eine effiziente Unterstützung beim Füllen von Beweislücken zur Verfügung stehen. Das im folgenden vorgestellte $\Omega$-Mkrp-System ist der Versuch einer Synthese der Ans?tze des vollautomatischen, des interaktiven und des planbasierten Beweisens. Dieser Artikel soll eine übersicht über unsere Arbeit an diesem System geben. Eingegangen am 24. Juni 1994 / Angenommen am 3. November 1995     

Recruiting Principals to Lead Low-performing Schools: Effects on Job Attractiveness

This study addressed recruiting principals into low performing schools. A sample (N = 619) of principals, assistant principals, central office administrators, school counselors, and teacher-leaders from the 29th largest school district in the United States role-played as applicants for principal vacancies at schools classified as low-performing based on results of standardized student achievement tests. The study had an experimental design and involved use of a recruitment simulation. Rating jobs depicted on simulated principal recruitment announcements, principals and assistant principals rated the job highest while school counselors rated the jobs lowest. Among the teacher-leaders, teacher-administrative interns rated the job higher than did highly skilled educators or resource teachers. Implications for recruitment practice and future evaluation research are discussed.     

Research Policy In Emerging Economies: Brazil’s Sector Funds

This paper examines Brazil’s Sector Funds, an emerging research and development policy that seeks to increase public-private research collaboration. The first section sets the context of the ‘policy problem’. The second section discusses the policy’s implementation, and discusses its likely implications for Brazil and beyond.