Scaling High-Quality Pairwise Link-Based Similarity Retrieval on Billion-Edge Graphs

SimRank is an attractive link-based similarity measure used in fertile fields of Web search and sociometry. However, the existing deterministic method by Kusumoto et al. [ 24 ] for retrieving SimRank does not always produce high-quality similarity results, as it fails to accurately obtain diagonal correction matrix D . Moreover, SimRank has a “connectivity trait” problem: increasing the number of paths between a pair of nodes would decrease its similarity score. The best-known remedy, SimRank++ [ 1 ], cannot completely fix this problem, since its score would still be zero if there are no common in-neighbors between two nodes. In this article, we study fast high-quality link-based similarity search on billion-scale graphs. (1) We first devise a “varied- D ” method to accurately compute SimRank in linear memory. We also aggregate duplicate computations, which reduces the time of [ 24 ] from quadratic to linear in the number of iterations. (2) We propose a novel “cosine-based” SimRank model to circumvent the “connectivity trait” problem. (3) To substantially speed up the partial-pairs “cosine-based” SimRank search on large graphs, we devise an efficient dimensionality reduction algorithm, PSR # , with guaranteed accuracy. (4) We give mathematical insights to the semantic difference between SimRank and its variant, and correct an argument in [ 24 ] that “if D is replaced by a scaled identity matrix (1-Ɣ)I, their top-K rankings will not be affected much”. (5) We propose a novel method that can accurately convert from Li et al. SimRank ~{S} to Jeh and Widom’s SimRank S . (6) We propose GSR # , a generalisation of our “cosine-based” SimRank model, to quantify pairwise similarities across two distinct graphs, unlike SimRank that would assess nodes across two graphs as completely dissimilar. Extensive experiments on various datasets demonstrate the superiority of our proposed approaches in terms of high search quality, computational efficiency, accuracy, and scalability on billion-edge graphs.

Anytime Ranking on Document-Ordered Indexes

Inverted indexes continue to be a mainstay of text search engines, allowing efficient querying of large document collections. While there are a number of possible organizations, document-ordered indexes are the most common, since they are amenable to various query types, support index updates, and allow for efficient dynamic pruning operations. One disadvantage with document-ordered indexes is that high-scoring documents can be distributed across the document identifier space, meaning that index traversal algorithms that terminate early might put search effectiveness at risk. The alternative is impact-ordered indexes, which primarily support top- disjunctions but also allow for anytime query processing, where the search can be terminated at any time, with search quality improving as processing latency increases. Anytime query processing can be used to effectively reduce high-percentile tail latency that is essential for operational scenarios in which a service level agreement (SLA) imposes response time requirements. In this work, we show how document-ordered indexes can be organized such that they can be queried in an anytime fashion, enabling strict latency control with effective early termination. Our experiments show that processing document-ordered topical segments selected by a simple score estimator outperforms existing anytime algorithms, and allows query runtimes to be accurately limited to comply with SLA requirements.

How to Optimize the Job Search Process: Development and Validation of the Job Search Quality Scale

Job search quality is important for unemployed individuals pursuing reemployment. To comprehensively measure job search quality, we develop and test a 20-item Job Search Quality Scale (JSQS), using four samples of unemployed individuals (pilot sample, N=218; exploration sample, N=3372; confirmation sample, N=3372; and replication sample, N=434). Results show a four-dimensional structure, composed of (a) goal establishment and planning, (b) preparation and alignment, (c) emotion regulation and persistence, and (d) learning and improvement. Substantial evidence was found for its reliability, convergent and discriminant validity. Building job search quality’s nomological net, conscientiousness, learning goal orientation, self-efficacy, employment commitment, autonomous job search motivation, and social support emerged as positive correlates. Supporting its criterion-related validity, the JSQS predicted key job search and employment outcomes. Moreover, usefulness analyses supported its incremental validity beyond extant job search measures. Our findings have important implications for studying and measuring job search quality in future research and career counseling practice.

The effect of librarian involvement on the quality of systematic reviews in dental medicine

Objectives To determine whether librarian or information specialist authorship is associated with better reproducibility of the search, at least three databases searched, and better reporting quality in dental systematic reviews (SRs). Methods SRs from the top ten dental research journals (as determined by Journal Citation Reports and Scimago) were reviewed for search quality and reproducibility by independent reviewers using two Qualtrics survey instruments. Data was reviewed for all SRs based on reproducibility and librarian participation and further reviewed for search quality of reproducible searches. Results Librarians were co-authors in only 2.5% of the 913 included SRs and librarians were mentioned or acknowledged in only 9% of included SRs. Librarian coauthors were associated with more reproducible searches, higher search quality, and at least three databases searched. Although the results indicate librarians are associated with improved SR quality, due to the small number of SRs that included a librarian, results were not statistically significant. Conclusion Despite guidance from organizations that produce SR guidelines recommending the inclusion of a librarian or information specialist on the review team, and despite evidence showing that librarians improve the reproducibility of searches and the reporting of methodology in SRs, librarians are not being included in SRs in the field of dental medicine. The authors of this review recommend the inclusion of a librarian on SR teams in dental medicine and other fields.

Upskilling health and care workers with augmented and virtual reality: protocol for a realist review to develop an evidence-informed programme theory

IntroductionAugmented reality (AR) and virtual reality (VR) are increasingly used to upskill health and care providers, including in surgical, nursing and acute care settings. Many studies have used AR/VR to deliver training, providing mixed evidence on their effectiveness and limited evidence regarding contextual factors that influence effectiveness and implementation. This review will develop, test and refine an evidence-informed programme theory on what facilitates or constrains the implementation of AR or VR programmes in health and care settings and understand how, for whom and to what extent they ‘work’.Methods and analysisThis realist review adheres to the Realist And Meta-narrative Evidence Syntheses: Evolving Standards (RAMESES) standards and will be conducted in three steps: theory elicitation, theory testing and theory refinement. First, a search will identify practitioner, academic and learning and technology adoption theories from databases (MEDLINE, Scopus, CINAHL, Embase, Education Resources Information Center, PsycINFO and Web of Science), practitioner journals, snowballing and grey literature. Information regarding contexts, mechanisms and outcomes will be extracted. A narrative synthesis will determine overlapping configurations and form an initial theory. Second, the theory will be tested using empirical evidence located from the above databases and identified from the first search. Quality will be assessed using the Mixed Methods Appraisal Tool (MMAT), and relevant information will be extracted into a coding sheet. Third, the extracted information will be compared with the initial programme theory, with differences helping to make refinements. Findings will be presented as a narrative summary, and the MMAT will determine our confidence in each configuration.Ethics and disseminationEthics approval is not required. This review will develop an evidence-informed programme theory. The results will inform and support AR/VR interventions from clinical educators, healthcare providers and software developers. Upskilling through AR/VR learning interventions may improve quality of care and promote evidence-based practice and continued learning. Findings will be disseminated through conference presentations and peer-reviewed journal articles.

Making the Invisible Visible – Techniques for Recovering Deleted SQLite Data Records

Forensic analysis and evidence collection for web browser activity is a recurring problem in digital investigation. It is not unusual for a suspect to cover his traces. Accordingly, the recovery of previously deleted data such as web cookies and browser history are important. Fortunately, many browsers and thousands of apps used the same database system to store their data: SQLite. Reason enough to take a closer look at this product. In this article, we follow the question of how deleted content can be made visible again in an SQLite-database. For this purpose, the technical background of the problem will be examined first. Techniques are presented with which it is possible to carve and recover deleted data records from a database on a binary level. A novel software solution called FQLite is presented that implements the proposed algorithms. The search quality, as well as the performance of the program, is tested using the standard forensic corpus. The results of a performance study are discussed, as well. The article ends with a summary and identifies further research questions.

Principle-Based Concept Analysis Methodology Using a Phased Approach With Quality Criteria

The study aimed to provide a detailed description of a process to conduct a phased principle-based concept analysis and to introduce quality criteria assessment for a phased principle-based concept analysis. Concept analysis explores how a concept is described, used and measured in the literature. This conceptual understanding is important to guide translational research to direct the development of evidence-based practice. The principle-based concept analysis is one approach of concept analysis used in published work, but the literature is lacking in articles clearly describing how to conduct it in practice. This article provides a methodology utilising a phased approach and by advancing on previous work; this approach includes a combination of a systematic search, quality criteria and qualitative analysis with principle-based concept analysis. Quality criteria for a phased principle-based concept analysis is introduced to critically assess articles against the four principles: epistemology, pragmatic, linguistic and logical. These improvements to the methodology promote transparency, rigour and replicability. This comprehensive systematic approach will aid future phased principle-based concept analyses and enable future comparisons of concept development, advancement and related concepts to improve the evidence base.