Neural information retrieval: at the end of the early years

A recent “third wave” of neural network (NN) approaches now delivers state-of-the-art performance in many machine learning tasks, spanning speech recognition, computer vision, and natural language processing. Because these modern NNs often comprise multiple interconnected layers, work in this area is often referred to as deep learning. Recent years have witnessed an explosive growth of research into NN-based approaches to information retrieval (IR). A significant body of work has now been created. In this paper, we survey the current landscape of Neural IR research, paying special attention to the use of learned distributed representations of textual units. We highlight the successes of neural IR thus far, catalog obstacles to its wider adoption, and suggest potentially promising directions for future research.     

How do elite ski jumpers handle the dynamic conditions in imitation jumps?

We examined the effect of boundary conditions in imitation ski jumping on movement dynamics and coordination. We compared imitation ski jumps with – and without – the possibility to generate shear propulsion forces. Six elite ski jumpers performed imitation jumps by jumping from a fixed surface and from a rolling platform. The ground reaction force vector, kinematics of body segments, and EMG of eight lower limb muscles were recorded. Net joint dynamics were calculated using inverse dynamics. The two imitation jumps differed considerably from each other with regard to the dynamics (moments, forces), whereas the kinematics were very similar. Knee power was higher and hip power was lower on the rolling platform than on the fixed surface. Mean EMG levels were very similar for both conditions, but differences in the development of muscle activity were indicated for seven of eight muscles. These differences are reflected in a subtle difference of the alignment of ground reaction force with centre of mass: the ground reaction force runs continuously close to but behind the centre of mass on the rolling platform and fluctuates around it on the fixed surface. This likely reflects a different strategy for controlling angular momentum.     

THE EDUCATIONAL MEANING OF COMMUNAL LAUGHTER: ON THE EXPERIENCE OF CORPOREAL DEMOCRACY

In this article Joris Vlieghe, Maarten Simons, and Jan Masschelein attempt to articulate a new way of dealing with the public character of education. Instead of discussing laughter as an instrument that one could use to facilitate established educational goals, the authors provide an extensive analysis of the phenomenon of laughter as a specific form of corporeal behavior. Their analysis demonstrates that when we laugh, we give an answer to a disorienting situation, but this answer is not the product of intentional agency. Instead, it consists in the uncontrollable spasmodic contraction of our diaphragm and other impersonal and automatic corporeal reactions. We are thus exposed to an ultimate loss of self‐control. The authors argue further that communal laughter—that is, when the lack of mastery over our own lives becomes a shared experience—results in what they call a “democracy of the flesh.” In this state, it is no longer possible to stick to well‐defined positions, nor is it tenable to defend any hierarchical ordering, including the strict hierarchy we typically find in the context of schooling and education. Common laughter makes equals out of us and grants the possibility of an unforeseen and unimaginable future. For precisely this reason, the authors conclude, communal laughter might be considered as an educational event itself.     

Transitions to post-secondary and tertiary education in the Netherlands: a trend analysis of unconditional and conditional socio-economic background effects

In the tracked educational system of the Netherlands, students at the end of secondary education have to decide whether they want to enter subsequent post-secondary or tertiary education. Depending on the previous qualification, they have the choice between up to four different options, including not entering further education. We propose, in line with prevalent theoretical approaches, that children from lower socio-economic backgrounds tend to make decisions that do not fully capitalize on their previously obtained qualifications. By means of multinomial logistic regression models we tested the unconditional and conditional effects of family background for entering the different tracks of post-secondary education. In the unconditional analyses we found effects for parental education on making a transition to all types of post-secondary and tertiary education, but the occupational status of the father seems to be only relevant for the transition to lower tier tertiary education. The conditional effects of parental education for making the transition to senior vocational education and university are strong, while the transition to lower tertiary education is not influenced by parental background characteristics. This shows that even with eligibility for the most prestigious tracks, children from lower socio-economic backgrounds tend to make less ambitious educational decisions. We also examined to what extent this inequality changed across time for the cohorts that terminated secondary education between 1932 and 1995. Decreasing effects of parental education indicate that the expansion of secondary education had the positive effect of leading more children from lower social backgrounds into favourable secondary education tracks, especially the intermediate general track. This equalization is carried forward through the entire sequence of educational transitions. The students from advantaged backgrounds nevertheless still profit from the parental resources in access to the most prestigious tertiary education institutions given secondary qualifications. Their head start into the academic track has not been reduced across cohorts.     

Success and failure in secondary education: socio‐economic background effects on secondary school outcome in the Netherlands, 1927–1998

In the Netherlands, educational attainment is the result of a sequence of separate educational transitions. Because of the tracked nature of the Dutch educational system, students do not make binary stay‐or‐leave‐decisions at each transition. After having entered one track of secondary education, students can change tracks during the entire secondary course. The initial track and the secondary school outcome therefore are incongruent for a significant proportion of the Dutch students. As social background partly predicts initial track placement, track changes and successful termination of the course, we suggest distinguishing conditional and unconditional effects of family background in the transition to secondary school outcome. This paper complements findings of previous research by taking into account the tracked structure of the Dutch educational system and the entire sequence of transitions in secondary education. For the empirical analysis, repeated cross‐sections from the Family Survey Dutch Population (1992, 1998, 2000 and 2003) are used. Multinomial logistic regressions reveal that inequality in the outcome of secondary education is partly explained by the fact that initial track placement is socially selective and because this initial inequality is even enhanced by track changes during secondary education. The remaining ‘conditional’ effect of parental education, however, indicates that parental education works on top of this selection to prevent drop out. Inequality in secondary school outcome thus is a cumulative result of social background effects in a sequence of educational transitions throughout secondary education. Decreasing inequality over time is entirely explained by decreasing inequality in the transition from primary to secondary education.     

The prevalence and risk indicators of symptoms of common mental disorders among current and former Dutch elite athletes

The aim of the study was to determine the prevalence and comorbidity of symptoms of common mental disorders (distress, anxiety/depression, sleep disturbance, eating disorders, adverse alcohol use) among current and former Dutch elite athletes, and to explore the inference between potential risk indicators (severe injury, surgery, life events, sport career dissatisfaction, social support) and the outcomes measures under investigation. Cross-sectional analyses were conducted on baseline questionnaires from an ongoing prospective cohort study among 203 current and 282 former elite Dutch athletes (response rate: 28% among current athletes and 95% among former athletes). Based on validated scales, an electronic questionnaire was set up and distributed. Prevalence (4-week) ranged from 6% for adverse alcohol use to 45% for anxiety/depression among current elite athletes, and from 18% for distress to 29% for anxiety/depression among former elite athletes. A higher number of past severe injuries, higher number of past surgeries, higher number of recent life events, higher level of career dissatisfaction and lower level of social support were related to the occurrence of symptoms of common mental disorders among both current and former elite athletes. On average, the 4-week prevalence of common mental disorders as shown in our study among current and former Dutch elite athletes were similar to the ones found among athletes from other sports disciplines and does compare with the lifetime prevalence estimates in the general population of the Netherlands.     

Explaining international sporting success: An international comparison of elite sport systems and policies in six countries

The aim of this article is to explore the relationship between elite sport policy systems (inputs and throughputs) and success in international competitions (outputs). A conceptual model of the sports policy factors, which lead to international sporting success, was implemented in an empirical environment in a pilot study with six nations. The study has sought to operationalise nine pillars, or key drivers in elite sport systems, into measurable concepts, which can be aggregated into an overall score for each pillar. In addition to a national sport policy questionnaire, athletes, coaches and performance directors were also involved in the collection of qualitative and quantitative data. Although the results are inconclusive, the findings suggest that some pillars could be regarded as possible drivers of an effective system because they were prioritised in the most successful sample nations: financial resources (pillar 1), athletic and post-career support (pillar 5), training facilities (pillar 6) and coach development (partly pillar 7).     

The Importance of Length Normalization for XML Retrieval

XML retrieval is a departure from standard document retrieval in which each individual XML element, ranging from italicized words or phrases to full blown articles, is a retrievable unit. The distribution of XML element lengths is unlike what we usually observe in standard document collections, prompting us to revisit the issue of document length normalization. We perform a comparative analysis of arbitrary elements versus relevant elements, and show the importance of element length as a parameter for XML retrieval. Within the language modeling framework, we investigate a range of techniques that deal with length either directly or indirectly. We observe a length-bias introduced by the amount of smoothing, and show the importance of extreme length bias for XML retrieval. We also show that simply removing shorter elements from the index (by introducing a cut-off value) does not create an appropriate element length normalization. Even after restricting the minimal size of XML elements occurring in the index, the importance of an extreme explicit length bias remains.