Automatic search from streaming data

Streaming data poses a variety of new and interesting challenges for information retrieval and text analysis. Unlike static document collections, which are typically analyzed and indexed off-line to support ad-hoc queries, streaming data often must be analyzed on the fly and acted on as the data passes through the analysis system. Speech is one example of streaming data that is a challenge to exploit, yet has significant potential to provide value in a knowledge management system. We are specifically interested in techniques that analyze streaming data and automatically find collateral information, or information that clarifies, expands, and generally enhances the value of the streaming data. We present a system that analyzes a data stream and automatically finds documents related to the current topic of discussion in the data stream. Experimental results show that the system generates result lists with an average precision at 10 hits of better than 60%. We also present a hit-list re-ranking technique based on named entity analysis and automatic text categorization that can improve the search results by 6%–12%.     

Can Legislation Reduce Gender Differences in Subject Choice? A Survey of GCSE and A level Entries Between 1970 and 1995

This paper investigates the extent of gender differences in subject choice patterns at GCSE and A level between 1970 and 1995. A Gender Inequality Index (GII) is derived to analyse national examination entry data, in particular to compare the impact of the Sex Discrimination Act and National Curriculum on the overall time trends. The paper finds that the GII slopes downwards over time, but that neither intervention had a significant impact on the rate of decline at A level and only the National Curriculum did so at GCSE level.     

Swings and roundabouts: Centralisation and devolution in a multi-campus university in South Africa

Over the course of its ninety-yearhistory a great deal of time and energy has beendevoted at the University of Natal to trying tounderstand the best way to administer and manage aunitary University whose two main campuses are fiftymiles apart. In the early 1990s, after decades ofgradual separation of the campuses, the massivechanges taking place in South Africa, as experiencedin Higher Education, prompted a major review of themanagement, administrative and committee structureswithin the University. This review, whoserecommendations were accepted and implemented,advocated a formal process of administrative andbudgetary devolution to the two campuses in Durban andPietermaritzburg. Five years later, in the face offurther external pressures, a new review recommendedthat the devolution be reversed and a stronglycentralised structure be put in place. Thisrecommendation, in turn, was approved and implemented. This article outlines the reasoning behind both setsof recommendations and explores the pros and cons fora multi-campus university of both centralisation anddevolution in the light of the experience of auniversity which has gained first-hand knowledge ofboth alternatives in less than a decade.     

The effect of voluntary ventilation on acid-base responses to a Moo Duk Tkow form

Voluntary control of breathing is taught in Moo Duk Tkow in order to maximize force during striking, kicking, and blocking. Form, a formalized practice routine which incorporates these techniques, is a short predetermined sequence of arm and leg movements performed at high intensity which includes a number of pauses. The purpose of this study was to investigate the acid-base and lactate responses to a beginning level form, Ki Cho I. Nine subjects, five black belts and four brown belts, were asked to perform Ki Cho I at competitive intensity. Arterialized finger capillary blood was drawn prior to and immediately after the form for the determination of pH, bicarbonate, base excess, and PCO2 by use of the Siggaard-Anderson Nomogram. Heart rate was obtained by ECG, and lactate was measured by enzymatic assay. Statistically significant differences (p less than .05) were found for all variables. Mean pre and post values (+/- standard deviations) were: pH 7.39 +/- .04 to 7.34 +/- .03; base excess -.3 +/- 1.2 to -2.8 +/- 1.7 mEq/L; bicarbonate 23.7 +/- 1.0 to 21.8 +/- 1.4 mEq/L; PCO2 39.9 +/- 3.9 to 45.1 +/- 5.4 mmHg; lactate 1.51 +/- .71 to 3.23 +/- 1.56 mM; and heart rate 74.9 +/- 12.6 to 132.1 +/- 17.0 bpm. The mean form time was 17.2 +/- 1.8 s. It was concluded that respiratory compensation does not occur and that respiratory acidosis may contribute to the metabolic acidosis.