Progress in Recommender Systems Research: Crisis? What Crisis?

Scholars in algorithmic recommender systems research have developed a largely standardized scientific method, where progress is claimed by showing that a new algorithm outperforms existing ones on or more accuracy measures. In theory, reproducing and thereby verifying such improvements is easy, as it merely involves the execution of the experiment code on the same data. However, as recent work shows, the reported progress is often only virtual, because of a number of issues related to (i) a lack of reproducibility, (ii) technical and theoretical flaws, and (iii) scholarship practices that are strongly prone to researcher biases. As a result, several recent works could show that the latest published algorithms actually do not outperform existing methods when evaluated independently. Despite these issues, we currently see no signs of a crisis, where researchers re-think their scientific method, but rather a situation of stagnation, where researchers continue to focus on the same topics. In this paper, we discuss these issues, analyze their potential underlying reasons, and outline a set of guidelines to ensure progress in recommender systems research.

Diagnostic accuracy of internal jugular vein ultrasound in quantification of the central venous pressure for hemodialysis patients

Background Assessment of the central venous pressure (CVP) is an essential hemodynamic parameter for monitoring the dialyzing patients. Our objective of the present study is to investigate the accuracy of CVP measurement by internal jugular vein US in comparison to the direct measurement by the central venous catheters for hemodialysis patients. We included 106 patients; where their CVP was assessed in two different non invasive US methods (CVPni) separately and in combination and the obtained measurements were correlated to the invasive measurements (CVPi) by catheters. Results By method 1, there is a highly significant positive correlation between CVPni and CVPi (ρ < 0.001) and a Pearson correlation coefficient (r = 0.913 n = 93), and by method 2, there is also a highly significant positive correlation between the CVPni and CVPi in both groups (r = 0.832, 95%, n = 106, p < 0.001), 1.935 was the cut-off point for prediction of CVP ≥ 10cmH20. For differentiation between patients with CVP < 10cmH20 and ≥ 10cmH20, the accuracy measures (sensitivity, specificity, PPV, NPV, and overall accuracy) were 100%, 79.31%, 74.47%, 100%, and 87.10% by method 1, and were 91.11%, 85.48%, 82.00%, 92.98%, and 87.85% by method 2, while the combination of both methods had gained 88.57%, 89.66%, 83.78%, 92.86%, and 89.25%, respectively. Conclusion The US offered a reliable and non-invasive tool for monitoring CVP. The present study has a novelty of combining more than one US method and this had reported higher accuracy measures and outperformed the use of a single method.

The Role of Whole- Body Diffusion Imaging with Background Signal Suppression (DWBS) in comparison with PET/CT in assessment of lymphoma patients

Background Malignant lymphomas [Hodgkin lymphomas (HL) and non-Hodgkin lymphoma (NHL)] rank third in incidence, of all childhood cancers, Moreover, in adolescents (aged 15-19 years) the malignant lymphomas are the leading cause of cancer. Furthermore, as with many other cancers, the likelihood of an individual being diagnosed with lymphoma increases markedly with age, with the median age at diagnosis being 67 years. Aim of the Work To compare between F-18-FDG PET-CT and whole-body diffusion-weighted imaging MR protocol (DWIBS) for initial staging and post chemotherapy evaluation in patients with pathologically proven lymphoma (Hodgkin and Non-Hodgkin). Patients and Methods The study is conducted on 32 patients with pathologically proven lymphoma to perform 18-F-FDG PET/CT either for pre-treatment initial staging or for evaluation of response to chemotherapy. A total of 22 had HD (69%) and 10 had NHL (31%). Staging PET/CT and WB-MRI/DWIBS were done at time of diagnosis in 19 of the 32 patients (59.4%), where immediate post-therapy PET/CT and WB-MRI-DWIBS were performed 13 patients (40.6%). Accuracy measures were calculated for PET CT and DWIBS. Results Accuracy measures confirm the higher sensitivity and specificity of PET-CT over DWIBS. Results for F-18 FDG PET/CT were clearly superior with statistically higher sensitivity, specificity, accuracy, PPV & NPV (96.3%, 99.16%, 98.43%, 97.5% and 98.75%) compared to (83.95%, 97.91%, 94.38%, 93.15% and 94.74%) for DWIBS results respectively. Conclusion F-18-FDG PET-CT remains a cornerstone in the evaluation of malignant lymphoma patients; it has significant higher sensitivity, specificity and overall accuracy compared to WB-MRI/DWIBS in HL and NHL patients, either in initial staging or in post therapy evaluation. WB-MRI/DWIBS in HL and NHL patients, either in initial staging or in post therapy evaluation. WB-MRI/DWIBS despite of its relatively longer acquisition time may provide a complementary tool for FDG PET CT.

An emotion-driven virtual counselling system in computer-mediated communication

We present and evaluate a virtual counselling system that is devoted to improving user awareness of emotional situations in computer-mediated communication and making informed decisions on actions to recommend to the users involved in a conversation. Starting from elements such as the moods and emotions of the users involved in a conversation, the system constructs the emotional signatures of individuals and groups that are used to characterize a situation. It then uses an approximate reasoning mechanism based on three-way decisions to classify recognized situations with respect to particular emotional dynamics based on emotional contagion. A prototype of the system has been experimented on in a real context based on collaboration between university students for the realization of project work. The distinctive features of the system have been evaluated with accuracy measures, and the results are promising.

Generating Up-to-Date Crop Maps Optimized for Sentinel-2 Imagery in Israel

The overarching aim of this research was to develop a method for deriving crop maps from a time series of Sentinel-2 images between 2017 and 2018 to address global challenges in agriculture and food security. This study is the first step towards improving crop mapping based on phenological features retrieved from an object-based time series on a national scale. Five main crops in Israel were classified: wheat, barley, cotton, carrot, and chickpea. To optimize the object-based classification process, different characteristics and inputs of the mean shift segmentation algorithm were tested, including vegetation indices, three-band combinations, and high/low emphasis on the spatial and spectral characteristics. Four known vegetation indices (VIs)-based time series were tested. Additionally, we compared two widely used machine learning methods for crop classification, support vector machine (SVM) and random forest (RF), in addition to a newer classifier, extreme gradient boosting (XGBoost). Lastly, we examined two accuracy measures—overall accuracy (OA) and area under the curve (AUC)—in order to optimally estimate the accuracy in the case of imbalanced class representation. Mean shift best performed when emphasizing both the spectral and spatial characteristics while using the green, red, and near-infrared (NIR) bands as input. Both accuracy measures showed that RF and XGBoost classified different types of crops with significantly greater success than achieved by SVM. Nevertheless, AUC was better able to represent the significant differences between the classification algorithms than OA was. None of the VIs showed a significantly higher contribution to the classification. However, normalized difference infrared index (NDII) with XGBoost classifier showed the highest AUC results (88%). This study demonstrates that the short-wave infrared (SWIR) band with XGBoost improves crop type classification results. Furthermore, the study emphasizes the importance of addressing imbalanced classification datasets by using a proper accuracy measure. Since object-based classification and phenological features derived from a VI-based time series are widely used to produce crop maps, the current study is also relevant for operational agricultural management and informatics at large scales.

Evaluation of the Cepheid Xpert C. difficile diagnostic assay: an update meta-analysis

Background Accurate and rapid diagnosis of Clostridium difficile infection (CDI) is critical for effective patient management and implementation of infection control measures to prevent transmission. Objectives We updated our previous meta-analysis to provide a more reliable evidence base for the clinical diagnosis of Xpert C. difficile (Xpert C. difficile) assay. Methods We searched PubMed, EMBASE, Cochrane Library, Chinese National Knowledge Infrastructure (CNKI), and the Chinese Biomedical Literature Database (CBM) databases to identify studies according to predetermined criteria. STATA 13.0 software was used to analyze the tests for sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, diagnostic odds ratio, and area under the summary receiver operating characteristic curves (AUC). QUADAS-2 was used to assess the quality of included studies with RevMan 5.2. Heterogeneity in accuracy measures was tested with Spearman correlation coefficient and chi-square. Meta-regressions and subgroup analyses were performed to figure out the potential sources of heterogeneity. Model diagnostics were used to evaluate the veracity of the data. Results A total of 26 studies were included in the meta-analysis. The pooled sensitivity (95% confidence intervals [CI]) for diagnosis was 0.97(0.95–0.98), and specificity was 0.96(0.95–0.97). The AUC was 0.99 (0.98–1.00). Model diagnostics confirmed the robustness of our meta-analysis’s results. Significant heterogeneity was still observed when we pooled most of the accuracy measures of selected studies. Meta-regression and subgroup analyses showed that the sample size and type, ethnicity, and disease prevalence might be the conspicuous sources of heterogeneity. Conclusions The up-to-date meta-analysis showed the Xpert CD assay had good accuracy for detecting CDI. However, the diagnosis of CDI must combine clinical presentation with diagnostic testing to better answer the question of whether the patient actually has CDI in the future, and inclusion of preanalytical parameters and clinical outcomes in study design would provide a more objective evidence base.

Various Topologies Generated from E j -Neighbourhoods via Ideals

One of the considerable subjects in mathematics is the study of topology. Deducing topology from arbitrary binary relations has enticed the attention of many researchers. So, we devote this article to generate some kinds of topologies from ideals and E j -neighborhoods which are induced from any binary relation. We define new types of approximations and accuracy measures from these topologies and then compare them with their counterparts induced directly from E j -neighborhoods and ideals. Also, we show that the approximations and accuracy measures given, herein, are better than those introduced in some previous studies under any arbitrary relation.