Major Outliers Research Articles

Comparative study of kinship network community detection approaches

This study employs five community detection methods on a rooted tree network with 600 nodes and 599 edges based on the Galo tribe’s kin naming system in Arunachal Pradesh, India. The network was originally assumed to be made up of three communities, which the algorithms were able to further divide into smaller groupings. The Louvain approach produced the most balanced distribution of community sizes, with skewness and kurtosis values close to zero, implying that the detected communities were reasonably evenly distributed in size and without major outliers. However, the Louvain algorithm found 25 communities, which is more than the network’s previously reported three communities. Further investigation may be required to integrate some of these communities in order to obtain the original known communities. Overall, this study highlights the importance of selecting an appropriate community detection algorithm for a given network and research question.

Gamson going global? Cabinet proportionality in comparative perspective

Abstract We conduct a global, large-N analysis of proportionality in the partisan distribution of cabinet portfolios. Formulated in the context of postwar Western European parliamentary democracy, Gamson’s Law predicts that parties joining a coalition government will receive cabinet ministries in direct proportion to the seats they are contributing to the coalition on the floor of the legislature. Using a sample of 1551 country-years of coalitional government in 97 countries from 1966 to 2019, and comparing all main constitutional formats (parliamentary, presidential, and semi-presidential), we find that Gamson’s Law does not travel well outside its context of origin. Among the constitutional predictors of cabinet proportionality, we find that pure presidentialism is a major outlier, with an exaggerated form of formateur advantage. Introducing party-system and assembly-level predictors to the debate, we find that party institutionalization tends to increase fairness in portfolio allocation within parliamentary systems only.

Infant Outcomes Categorized by Birth Weight Percentile for Deliveries Between 28 and 41 Weeks of Gestation.

To examine the association between birth weight percentile and severe infant outcomes according to gestational age category. We conducted a population-based retrospective cohort study using publicly available U.S. birth certificate data linked to infant death data from 2017 to 2019. Maternal-neonate pairs of singleton live births between 28 0/7 and 41 6/7 weeks of gestation (vaginal or cesarean delivery) were evaluated. We excluded infants with major fetal anomalies, chromosomal disorders, missing data, and birth weight outliers. The primary outcome was infant mortality within 1 year of life. Neonates were divided into eight sex-specific birth weight percentile categories: less than the 3rd, 3rd-less than the 10th, 10th-less than the 25th, 25th-to less than the 49th, 50th-less than the 75th, 75th-to less than the 90th, 90th-less than the 97th, and 97th or higher. There were 10,459,388 births between 28 and 41 weeks of gestation: 69,793 (0.7%) at 28-31 weeks, 88,673 (0.8%) at 32-33 weeks, 635,904 (6.1%) at 34-36 weeks, 2,763,110 (26.4%) at 37-38 weeks, 6,269,894 (59.9%) at 39-40 weeks, and 632,014 (6.0%) at 41 weeks. Infant mortality during the first year of life significantly increased at higher and lower birth weight percentiles at 28-36 weeks of gestation, with the lowest risk observed at the 50th-less than the 75th percentile. The highest mortality rates were 13.6% at less than the 3rd percentile and 8.4% at the 97th percentile or higher at 28-31 weeks of gestation; the second-highest mortality rates were 7.7% at less than the 3rd percentile and 3.1% at the 97th percentile or higher at 32-33 weeks of gestation. At 34-36 weeks of gestation, the highest and second-highest mortality rates were 3.4% at less than the 3rd percentile and 1.4% at the 3rd-10th percentile. At 37-41 weeks of gestation, infant mortality was associated with lower birth weight percentile, but higher birth weight percentiles were not significantly associated with increased mortality. We found different patterns in the association between birth weight percentile and infant mortality depending on gestational age category.

Subject-Specific Automatic Reconstruction of White Matter Tracts

MRI-based tractography is still underexploited and unsuited for routine use in brain tumor surgery due to heterogeneity of methods and functional–anatomical definitions and above all, the lack of a turn-key system. Standardization of methods is therefore desirable, whereby an objective and reliable approach is a prerequisite before the results of any automated procedure can subsequently be validated and used in neurosurgical practice. In this work, we evaluated these preliminary but necessary steps in healthy volunteers. Specifically, we evaluated the robustness and reliability (i.e., test–retest reproducibility) of tractography results of six clinically relevant white matter tracts by using healthy volunteer data (N = 136) from the Human Connectome Project consortium. A deep learning convolutional network-based approach was used for individualized segmentation of regions of interest, combined with an evidence-based tractography protocol and appropriate post-tractography filtering. Robustness was evaluated by estimating the consistency of tractography probability maps, i.e., averaged tractograms in normalized space, through the use of a hold-out cross-validation approach. No major outliers were found, indicating a high robustness of the tractography results. Reliability was evaluated at the individual level. First by examining the overlap of tractograms that resulted from repeatedly processed identical MRI scans (N = 10, 10 iterations) to establish an upper limit of reliability of the pipeline. Second, by examining the overlap for subjects that were scanned twice at different time points (N = 40). Both analyses indicated high reliability, with the second analysis showing a reliability near the upper limit. The robust and reliable subject-specific generation of white matter tracts in healthy subjects holds promise for future validation of our pipeline in a clinical population and subsequent implementation in brain tumor surgery.

Understanding Internal Cluster Variability Through Subcluster Metric Analysis in a Geophysical Context

AbstractClustering algorithms are commonly used for inspecting the behavior of clouds in both model and satellite data sets. Often overlooked in cluster analysis is the variability that occurs within any clusters generated. This is particularly important in the geophysics where clusters are often generated with a focus on interpretability over mathematical optimization. Two metrics, the Davies‐Bouldin index and the subsom entropy, are used to identify clusters with large internal variability. These metrics are applied to an example set of clusters from prior research that were generated using cloud top pressure‐cloud optical thickness joint histograms from the Moderate Resolution Imaging Spectroradiometer data set. Applying these metrics to the clusters identifies one cluster in particular as a major outlier. Examining the calculations behind these metrics in more detail provides further information about the internal variability of the clusters. The clusters are also examined over several geographic regions showing mostly consistent behavior. There are, however, some large anomalies such as the behavior of the clear sky cluster or the behavior of several different clusters over the Arctic Ocean. To aide our interpretation of these results, two clusters are chosen for a detailed analysis of their subclusters. The geographic distributions and radiative properties of these subclusters are examined and clearly identify that subclusters have physically distinct behavior. This result illustrates that these metrics are capable of determining when a cluster contains physically distinct subclusters. This demonstrates the potential utility of these metrics if they were applied to other geophysical data sets.

An alternate proficiency testing tool for a point of care hemoglobinometer: Hemocue

Background: Pre-donation hemoglobin screening, to determine donor fitness, can be done via different modalities [from crude copper sulphate solutions to point-of-care (POC) hemoglobinometers to automated cell counters]. Primary aim of the study was to develop and implement strategies for quality control for a Hemocue (a POC device) for the pre – donation hemoglobin screening. Materials and Methods: A prospective observational study was done in properly stored 200 EDTA samples from April 2020 to October 2020. Daily two samples were analyzed using Hemocue and Automated hematology analyzer (Sysmex XN- 550) and compared by plotting their differences on Levey-Jennings (LJ) charts. Results: 100 samples were >12.5gm/dL and 100 samples were 12.5, 13.7± 0.75 for hemocue > 12.5 and 11.13 ± 0.96 for sysmex < 12.5, 11.37 ± 0.97 for hemocue < 12. No major outliers (differences outside 2 SD) were found in the LJ Plots. Conclusion: Sample rerun could be used as a safer alternative to running costly commercial controls for screening of donor with POC hemoglobinometers.

Irrigation with Coal Mining Effluents: Sustainability and Water Quality Considerations (São Pedro da Cova, North Portugal)

Two water effluents that drain from the abandoned coal mine of São Pedro da Cova (NW Portugal) were characterized in terms of their physic-chemical properties and suitability for irrigation purposes. Samples were also collected in a local surface stream, upstream and downstream from the mine drainage points, also used for irrigation by local farmers. Water samples were analyzed for major and minor ions and for trace element concentrations. Sampling campaigns started in 2017 and ended in 2019 and there were 46 water quality parameters tested. There were also proposed all-inclusive indices (the Water Quality Index and the Contamination Index, and also the Trace Element Toxicity Index) based on specific groups of 18 and 17 physic-chemical parameters, respectively, to achieve adequate monitoring requirements for mine effluents and surface water from coalfield. From the physical and chemical aspects of mine water it is inferred that the mine is not producing acid mine drainage. The coal mine water is of medium to high salinity, having almost neutral pH and a high thermal stability during the year, which is a distinguishing feature of the effluents. When compared to international irrigation water quality standards, as Food and Agriculture Organization of the United Nations admissible concentrations, the impacted waters are unsuitable for irrigation. The major outliers to the guidelines were iron, manganese, potassium, magnesium and bicarbonates, being also detected carcinogenic polycyclic aromatic hydrocarbons. Cost-effective ways of monitoring water quality parameters are needed to help control and manage the impact of coal mine effluents that should be treated before releasing into a ditch system that could be then used by local farmers to irrigate their crops.

Regional Solar Irradiance Forecast for Kanto Region by Support Vector Regression Using Forecast of Meso-Ensemble Prediction System

From the perspective of stable operation of the power transmission system, the transmission system operators (TSO) needs to procure reserve adjustment power at the stage of the previous day based on solar power forecast information from global horizontal irradiance (GHI). Because the reserve adjustment power is determined based on information on major outliers in past forecasts, reducing the maximum forecast error in addition to improving the average forecast accuracy is extremely important from the perspective of grid operation. In the past, researchers have proposed various methods combining the numerical weather prediction (NWP) and machine learning techniques for the one day-ahead solar power forecasting, but the accuracy of NWP has been a bottleneck issue. In recent years, the development of the ensemble prediction system (EPS) forecasts based on probabilistic approaches has been promoted to improve the accuracy of NWP, and in Japan, EPS forecasts in the mesoscale domain, called mesoscale ensemble prediction system (MEPS), have been distributed by the Japan Meteorological Agency (JMA). The use of EPS as a machine learning model is expected to improve the maximum forecast error, as well as the accuracy, since the predictor can utilize various weather scenarios as information. The purpose of this study is to examine the effect of EPS on the GHI prediction and the structure of the machine learning model that can effectively use EPS. In this study, we constructed the support vector regression (SVR)-based predictors with multiple network configurations using MEPS as input and evaluated the forecast error of the Kanto region GHI by each model. Through the comparison of the prediction results, it was shown that the machine learning model can achieve average accuracy improvement while reducing the maximum prediction error by MEPS, and knowledge was obtained on how to effectively provide EPS information to the predictor. In addition, machine learning was found to be useful in improving the systematic error of MEPS.

Assessing European Union standardization: a descriptive analysis of voluntary ex ante transparency notices

PurposeThis study aims to empirically assess the standardization of using voluntary ex ante transparency notices to announce the awards of noncompetitive large-value contracts.Design/methodology/approachBased on open data published in theOfficial Journal of the European Union, a pooled cross-sectional research design is used to determine the level of standardized use of noncompetitive contracting by member states.FindingsFindings suggest little evidence of standardization when publicizing direct contract awards, which might warrant remedial measures for promoting standardization by the EU. Moreover, France was found to be a major outlier in the prevalence of using non-competitive direct contract awards procedures.Social implicationsMaintenance of the European Union is predicated on free, transparent and open competition among member states, and this can only be maintained if each member state transposes EU standards into their national laws.Originality/valueFindings suggest little evidence of standardization when publicizing direct contract awards, which might warrant additional remedial measures promoting consistency across the EU. Moreover, France was found to be a major outlier in the prevalence of using non-competitive direct contract awards procedures.

The NSW Pathology Atlas of Variation: Part I—Identifying Emergency Departments With Outlying Laboratory Test–Ordering Practices

Abdominal pain and chest pain are leading reasons for emergency department (ED) presentations, with laboratory tests frequently ordered to aid the diagnostic process. Our study aims to identify EDs with outlying laboratory ordering practices for patients presenting with undifferentiated abdominal pain and chest pain. This was a retrospective observational study of 519,597 patients who presented with the complaint of abdominal or chest pain at 44 major hospital EDs across New South Wales, Australia, from January 2017 to September 2018. For each condition, we evaluated the risk-adjusted rate of ordering at least 1 laboratory test and of each of the top 15 ordered tests. We used funnel plots to graph variations in test ordering and identify EDs with outlying test-ordering practices. EDs lying above or below the 99.8% funnel control limits were regarded as outliers. From 3,360,152 unplanned presentations, abdominal pain and chest pain represented 8.8% (n=296,809) and 6.6% (n=222,788) of all cases, respectively. No major outliers were observed for ordering at least one laboratory test; however, variations were observed for individual tests. For abdominal pain, the top 3 tests with the highest ordering variation included glucose (20 outlier EDs), C-reactive protein (10 outliers), and calcium-magnesium-phosphate (7 outliers). For chest pain, the top 3 tests with the highest ordering variation were glucose (21 outlier EDs), C-reactive protein (17 outliers), and liver function test (14 outliers). Identifying EDs with outlying laboratory-ordering practices is the first step in initiating context-specific evaluation of whether outlying variations are unwarranted.

A Recursive Subdivision Technique for Sampling Multi-class Scatterplots.

We present a non-uniform recursive sampling technique for multi-class scatterplots, with the specific goal of faithfully presenting relative data and class densities, while preserving major outliers in the plots. Our technique is based on a customized binary kd-tree, in which leaf nodes are created by recursively subdividing the underlying multi-class density map. By backtracking, we merge leaf nodes until they encompass points of all classes for our subsequently applied outlier-aware multi-class sampling strategy. A quantitative evaluation shows that our approach can better preserve outliers and at the same time relative densities in multi-class scatterplots compared to the previous approaches, several case studies demonstrate the effectiveness of our approach in exploring complex and real world data.

Peer‐to‐peer fundraising shows modest improvement for most charities

Total revenues for the top 30 U.S. peer‐to‐peer fundraising campaigns declined slightly in 2018, but a deeper look at the numbers shows these events are actually on the upswing. As laid out in the 2018 Peer‐to‐Peer Fundraising Thirty, an annual survey by the Peer‐to‐Peer Professional Forum, the overall decrease was much less than the year before, and if you took out one major outlier that posted a huge decrease for the year, the numbers actually show a revenue increase for 2018, indicating that the overall climate for peer‐to‐peer fundraising is improving.

Network-based microsynteny analysis identifies major differences and genomic outliers in mammalian and angiosperm genomes

A comprehensive analysis of relative gene order, or microsynteny, can provide valuable information for understanding the evolutionary history of genes and genomes, and ultimately traits and species, across broad phylogenetic groups and divergence times. We have used our network-based phylogenomic synteny analysis pipeline to first analyze the overall patterns and major differences between 87 mammalian and 107 angiosperm genomes. These two important groups have both evolved and radiated over the last ∼170 MYR. Secondly, we identified the genomic outliers or "rebel genes" within each clade. We theorize that rebel genes potentially have influenced trait and lineage evolution. Microsynteny networks use genes as nodes and syntenic relationships between genes as edges. Networks were decomposed into clusters using the Infomap algorithm, followed by phylogenomic copy-number profiling of each cluster. The differences in syntenic properties of all annotated gene families, including BUSCO genes, between the two clades are striking: most genes are single copy and syntenic across mammalian genomes, whereas most genes are multicopy and/or have lineage-specific distributions for angiosperms. We propose microsynteny scores as an alternative and complementary metric to BUSCO for assessing genome assemblies. We further found that the rebel genes are different between the two groups: lineage-specific gene transpositions are unusual in mammals, whereas single-copy highly syntenic genes are rare for flowering plants. We illustrate several examples of mammalian transpositions, such as brain-development genes in primates, and syntenic conservation across angiosperms, such as single-copy genes related to photosynthesis. Future experimental work can test if these are indeed rebels with a cause.

Outlier Detection in Wearable Sensor Data for Human Activity Recognition (HAR) Based on DRNNs

Wearable sensors provide a user-friendly and non-intrusive mechanism to extract user-related data that paves the way to the development of personalized applications. Within those applications, human activity recognition (HAR) plays an important role in the characterization of the user context. Outlier detection methods focus on finding anomalous data samples that are likely to have been generated by a different mechanism. This paper combines outlier detection and HAR by introducing a novel algorithm that is able both to detect information from secondary activities inside the main activity and to extract data segments of a particular sub-activity from a different activity. Several machine learning algorithms have been previously used in the area of HAR based on the analysis of the time sequences generated by wearable sensors. Deep recurrent neural networks (DRNNs) have proven to be optimally adapted to the sequential characteristics of wearable sensor data in previous studies. A DRNN-based algorithm is proposed in this paper for outlier detection in HAR. The results are validated both for intra- and inter-subject cases and both for outlier detection and sub-activity recognition using two different datasets. A first dataset comprising 4 major activities (walking, running, climbing up, and down) from 15 users is used to train and validate the proposal. Intra-subject outlier detection is able to detect all major outliers in the walking activity in this dataset, while inter-subject outlier detection only fails for one participant executing the activity in a peculiar way. Sub-activity detection has been validated by finding out and extracting walking segments present in the other three activities in this dataset. A second dataset using four different users, a different setting and different sensor devices is used to assess the generalization of results.

How Crashes Develop: Intradaily Volatility and Crash Evolution

ABSTRACTThis paper explores whether affine models with volatility jumps estimated on intradaily S&amp;P 500 futures data over 1983 to 2008 can capture major daily outliers such as the 1987 stock market crash. Intradaily jumps in futures prices are typically small; self‐exciting but short‐lived volatility spikes capture intradaily and daily returns better. Multifactor models of the evolution of diffusive variance and jump intensities improve fits substantially, including out‐of‐sample over 2009 to 2016. The models capture reasonably well the conditional distributions of daily returns and realized variance outliers, but underpredict realized variance inliers. I also examine option pricing implications.

Visualizing the Paris Climate Talks on Twitter: Media and Climate Stakeholder Visual Social Media During COP21

In 2015, meeting in Paris for the Conference of the Parties (COP21), representatives of 195 nations set an ambitious goal to reach net zero greenhouse gas (GHG) emissions by mid-century. This research uses the Parties to the United Nations Framework Convention on Climate Change, which took place in Paris during 30 November to 11 December 2015, as a case study of Twitter coverage of the talks by mainstream and alternative media outlets and other climate stakeholders, including activists and fossil fuel industry groups. It compares the British Guardian with other media and climate stakeholders’ visual framing of climate change on Twitter during COP21, because the publication had launched an advocacy campaign in March 2015 promoting fossil fuel divestment in the lead-up to COP21. Findings show that individual activists and movement organizations functioned similarly in climate change visual framing in Twitter posts, as did individual and organizational multinational representatives and scientific experts. The news media categories varied by type of news organization. The major outliers were the fossil fuel industry and trade association accounts. Industry stakeholders largely focused on former US President Barack Obama’s climate policy, promoting the perception of a lack of domestic support for his climate policies in their visual Twitter postings.

Dialysate bicarbonate concentration: Too much of a good thing?

Acid-base equilibrium is a complex and vital system whose regulation is impaired in chronic kidney disease (CKD). Metabolic acidosis is a common complication of CKD. It is typically due to the accumulation of sulfate, phosphorus, and organic anions. Metabolic acidosis is correlated with several adverse outcomes, such as morbidity, hospitalization and mortality. In patients undergoing hemodialysis, acid-base homeostasis depends on many factors: net acid production, amount of alkali given by the dialysate bath, duration of interdialytic period, as well as residual diuresis, if any. Recent literature data suggest that the development of postdialysis metabolic alkalosis may contribute to adverse clinical outcomes. Unfortunately, no randomized studies exist about the effect of different dialysate bicarbonate concentrations on hard outcomes, such as mortality. Like everything else in dialysis, the quest for the "ideal" dialysate bicarbonate concentration is far from over. The Latin aphorism "ne quid nimis" ie "nothing in excess" (excess of neither acid nor base) probably best summarizes our current state of knowledge in this field. For the present, the clinician should understand that target values for predialysis serum bicarbonate concentrations have been established primarily based on observational studies and expert opinion. On the basis of this information, we should keep predialysis serum bicarbonate concentrations at least at 22mEq/L. Furthermore, a specific focus should be addressed to the clinical and nutritional status of the major outliers on both the acid and alkaline sides of the curve.

The Outer Halo of the Milky Way as Probed by RR Lyr Variables from the Palomar Transient Facility*

Abstract RR Lyrae stars are ideal massless tracers that can be used to study the total mass and dark matter content of the outer halo of the Milky Way (MW). This is because they are easy to find in the light-curve databases of large stellar surveys and their distances can be determined with only knowledge of the light curve. We present here a sample of 112 RR Lyr stars beyond 50 kpc in the outer halo of the MW, excluding the Sgr streams, for which we have obtained moderate-resolution spectra with Deimos on the Keck II Telescope. Four of these have distances exceeding 100 kpc. These were selected from a much larger set of 447 candidate RR Lyr stars that were data-mined using machine-learning techniques applied to the light curves of variable stars in the Palomar Transient Facility database. The observed radial velocities taken at the phase of the variable corresponding to the time of observation were converted to systemic radial velocities in the Galactic standard of rest. From our sample of 112 RR Lyr stars we determine the radial velocity dispersion in the outer halo of the MW to be ∼90 km s−1 at 50 kpc, falling to about 65 km s−1 near 100 kpc once a small number of major outliers are removed. With reasonable estimates of the completeness of our sample of 447 candidates and assuming a spherical halo, we find that the stellar density in the outer halo declines as .

Environmental impact of coal mining and coal seam gas production on surface water quality in the Sydney basin, Australia.

The extraction of coal and coal seam gas (CSG) will generate produced water that, if not adequately treated, will pollute surface and groundwater systems. In Australia, the discharge of produced water from coal mining and related activities is regulated by the state environment agency through a pollution licence. This licence sets the discharge limits for a range of analytes to protect the environment into which the produced water is discharged. This study reports on the impact of produced water from coal mine activities located within or discharging into high conservation environments, such as National Parks, in the outer region of Sydney, Australia. The water samples upstream and downstream from the discharge points from six mines were taken, and 110 parameters were tested. The results were assessed against a water quality index (WQI) which accounts for pH, turbidity, dissolved oxygen, biochemical oxygen demand, total dissolved solids, total phosphorus, nitrate nitrogen and E .coli. The water quality assessment based on the trace metal contents against various national maximum admissible concentration (MAC) and their corresponding environmental impacts was also included in the study which also established a base value of water quality for further study. The study revealed that impacted water downstream of the mine discharge points contained higher metal content than the upstream reference locations. In many cases, the downstream water was above the Australia and New Zealand Environment Conservation Council and international water quality guidelines for freshwater stream. The major outliers to the guidelines were aluminium (Al), iron (Fe), manganese (Mn), nickel (Ni) and zinc (Zn). The WQI of surface water at and downstream of the discharge point was lower when compared to upstream or reference conditions in the majority of cases. Toxicology indices of metals present in industrial discharges were used as an additional tool to assess water quality, and the newly proposed environmental water quality index (EWQI) lead to better trend in the impact of coal and coal seam gas mining activities on surface water quality when compared to the upstream reference water samples. Metal content limits were based on the impact points assigned by the Agency for Toxic Substances and Disease Registry, USA. For environmental and health impact assessment, the approach used in this study can be applied as a model to provide a basis to assess the anthropogenic contribution from the industrial and mining activities on the environment.

Observational Evidence for High Neutronization in Supernova Remnants: Implications for Type Ia Supernova Progenitors

Abstract The physical process whereby a carbon–oxygen white dwarf explodes as a Type Ia supernova (SN Ia) remains highly uncertain. The degree of neutronization in SN Ia ejecta holds clues to this process because it depends on the mass and the metallicity of the stellar progenitor, and on the thermodynamic history prior to the explosion. We report on a new method to determine ejecta neutronization using Ca and S lines in the X-ray spectra of Type Ia supernova remnants (SNRs). Applying this method to Suzaku data of Tycho, Kepler, 3C 397, and G337.2−0.7 in the Milky Way, and N103B in the Large Magellanic Cloud, we find that the neutronization of the ejecta in N103B is comparable to that of Tycho and Kepler, which suggests that progenitor metallicity is not the only source of neutronization in SNe Ia. We then use a grid of SN Ia explosion models to infer the metallicities of the stellar progenitors of our SNRs. The implied metallicities of 3C 397, G337.2−0.7, and N103B are major outliers compared to the local stellar metallicity distribution functions, indicating that progenitor metallicity can be ruled out as the origin of neutronization for these SNRs. Although the relationship between ejecta neutronization and equivalent progenitor metallicity is subject to uncertainties stemming from the 12C + 16O reaction rate, which affects the Ca/S mass ratio, our main results are not sensitive to these details.