Query reformulation mining: models,patterns, and applications

Understanding query reformulation patterns is a key task towards next generation web search engines. If we can do that, then we can build systems able to understand and possibly predict user intent, providing the needed assistance at the right time, and thus helping users locate information more effectively and improving their web-search experience. As a step in this direction, we build a very accurate model for classifying user query reformulations into broad classes (generalization, specialization, error correction or parallel move), achieving 92% accuracy. We then apply the model to automatically label two very large query logs sampled from different geographic areas, and containing a total of approximately 17 million query reformulations. We study the resulting reformulation patterns, matching some results from previous studies performed on smaller manually annotated datasets, and discovering new interesting reformulation patterns, including connections between reformulation types and topical categories. We annotate two large query-flow graphs with reformulation type information, and run several graph-characterization experiments on these graphs, extracting new insights about the relationships between the different query reformulation types. Finally we study query recommendations based on short random walks on the query-flow graphs. Our experiments show that these methods can match in precision, and often improve, recommendations based on query-click graphs, without the need of users’ clicks. Our experiments also show that it is important to consider transition-type labels on edges for having recommendations of good quality.     

Effectiveness of bibliographic searches performed by paediatric residents and interns assisted by librarians. A randomised controlled trial

Background: Considerable barriers still prevent paediatricians from successfully using information retrieval technology. Objectives: To verify whether the assistance of biomedical librarians significantly improves the outcomes of searches performed by paediatricians in biomedical databases using real‐life clinical scenarios. Methods: In a controlled trial at a paediatric teaching hospital, nine residents and interns were randomly allocated to an assisted search group and nine to a non‐assisted (control) group. Each participant searched Pub Med and other online sources, performing pre‐determined tasks including the formulation of a clinical question, retrieval and selection of bibliographic records. In the assisted group, participants were supported by a librarian with ≥5 years of experience. The primary outcome was the success of search sessions, scored against a specific assessment tool. Results: The median score of the assisted group was 73.6 points interquartile range (IQR = 13.4) vs. 50.4 (IQR = 17.1) of the control group. The difference between median values in the results was 23.2 points (95% CI 4.8–33.2), in favour of the assisted group (P‐value, Mann–Whitney U test: 0.013). Conclusions: The study has found quantitative evidence of a significant difference in search performance between paediatric residents or interns assisted by a librarian and those searching the literature alone.     

Science or graduates: How do firms benefit from the proximity of universities?

This paper examines the impact of universities on the technological performance of adjacent firms. We extend existing research by jointly analyzing, and comparing, the effects of education (graduates) and scientific research (publications) activities of universities on firms’ technological performance. Adopting the knowledge production framework, our study is conducted at the level of 101 Italian territorial areas (provinces) and four industries. Overall, fixed-effect panel data models reveal a positive effect of both university graduates and scientific publications on the technological performance of firms. At the same time, considerable industry differences are observed. While the provision of university graduates positively affects firm performance in all industries under study, additional effects for scientific research are only observed in electrical and pharmaceutical industries that are science-intensive and where the scientific knowledge base is changing rapidly over time. The observation that spillovers from academia into the industrial texture of provinces rely on education and research in an industry-specific manner is relevant to the design of appropriate research and innovation policies.     

Evaluating the Power of Latent Growth Curve Models to Detect Individual Differences in Change

We evaluated the statistical power of single-indicator latent growth curve models to detect individual differences in change (variances of latent slopes) as a function of sample size, number of longitudinal measurement occasions, and growth curve reliability. We recommend the 2 degree-of-freedom generalized test assessing loss of fit when both slope-related random effects, the slope variance and intercept-slope covariance, are fixed to 0. Statistical power to detect individual differences in change is low to moderate unless the residual error variance is low, sample size is large, and there are more than four measurement occasions. The generalized test has greater power than a specific test isolating the hypothesis of zero slope variance, except when the true slope variance is close to 0, and has uniformly superior power to a Wald test based on the estimated slope variance.     

Ginzburg in Harlem: History,Structure and the Politics of Primitivism

Abstract

The historian Carlo Ginzburg is renowned for his critique of modern, scientific reason and his articulation of an alternative form of knowledge which he labels ‘conjectural’. This form of knowledge, supposedly more attuned to the historian’s interest in the singular and specific fragment, as opposed to the abstract and universal concept, is so rooted in the practices of the prehistoric hunter that Ginzburg sometimes describes it as a ‘venatic’ form of deduction, binding ‘the human animal closely to other animal species’. In this essay, I explore the ramifications of this alternative form of knowledge, attending especially to its relationship to the modernist theme of ‘primitivism’. I do so by juxtaposing Ginzburg’s critical appraisal of Arthur Conan Doyle’s most famous literary invention, Sherlock Holmes, and Rudolph Fisher’s own literary invention, John Archer, the physician who sometimes aids criminal investigations in African American Harlem. I argue that the differences between Archer and Holmes draw attention to some troubling implications of Ginzburg’s historiographical argument. Folding this analysis on itself, however, I also suggest that what might be at stake, when Ginzburg insists so troublingly on the importance of the singular, venatic trace, is the evocation of Walter Benjamin’s understanding of the historical ‘event’.     

Using the Web as corpus for self-training text categorization

Most current methods for automatic text categorization are based on supervised learning techniques and, therefore, they face the problem of requiring a great number of training instances to construct an accurate classifier. In order to tackle this problem, this paper proposes a new semi-supervised method for text categorization, which considers the automatic extraction of unlabeled examples from the Web and the application of an enriched self-training approach for the construction of the classifier. This method, even though language independent, is more pertinent for scenarios where large sets of labeled resources do not exist. That, for instance, could be the case of several application domains in different non-English languages such as Spanish. The experimental evaluation of the method was carried out in three different tasks and in two different languages. The achieved results demonstrate the applicability and usefulness of the proposed method.     

Return mobility and scientific productivity of researchers working abroad: The role of home country linkages

Through an analysis of 497 foreign researchers in Italy and Portugal we verify the impact of home linkages on return mobility choices and scientific productivity. We consider the presence of several different types of linkages of the researchers working abroad with their country of origin and control for the most relevant contextual factors (age, research area, position in the host country, etc.). The probability of return to their home country and scientific productivity in the host country are both higher for researchers that maintain home linkages. We conclude that the presence of home linkages directly benefits both countries in addition to the indirect benefit of expanding the scientific networks. Policy implications and suggestions for further research are discussed.     

Petrographic and chemical characterisation of fine ware from three Archaic and Hellenistic kilns in Gela,Sicily

This work is focused on the petro-archeometric characterization of the fine pottery production of the archaeological site of Gela (Sicily). Thirty-five samples coming from three Archaic and Hellenistic kilns and five coming from an Hellenistic house, were investigated by means of optical microscopy, XRD and XRF analysis. With the exception of some ceramics which are undoubtedly imported, all the samples have similar petrographic features, but on the basis of chemical composition, we can distinguish two different groups. The presence of two local productions is confirmed by the comparison with locally outcropping sediments and with archaic and modern bricks of sure Geloan production. This result is particularly significant for the definition of reference groups of this wealthy polis that played an important trading role in the Mediterranean area.