We study fragmentation in electron-positron annihilation assuming a dijet situation, using variables defined independent of any frame. In a collinear situation some of the variables are centered ...around zero with the small deviations attributed to intrinsic transverse momenta and large deviations attributed to additional hard subprocesses. Of course there is a gradual transition. Our modest goal is to show that covariantly defined variables are well suited to get a feeling for the magnitude of intrinsic transverse momenta.
RUSBoost: A Hybrid Approach to Alleviating Class Imbalance Seiffert, C.; Khoshgoftaar, T.M.; Van Hulse, J. ...
IEEE transactions on systems, man and cybernetics. Part A, Systems and humans,
2010-Jan., 2010-01-00, 20100101, Letnik:
40, Številka:
1
Journal Article
Odprti dostop
Class imbalance is a problem that is common to many application domains. When examples of one class in a training data set vastly outnumber examples of the other class(es), traditional data mining ...algorithms tend to create suboptimal classification models. Several techniques have been used to alleviate the problem of class imbalance, including data sampling and boosting. In this paper, we present a new hybrid sampling/boosting algorithm, called RUSBoost, for learning from skewed training data. This algorithm provides a simpler and faster alternative to SMOTEBoost, which is another algorithm that combines boosting and data sampling. This paper evaluates the performances of RUSBoost and SMOTEBoost, as well as their individual components (random undersampling, synthetic minority oversampling technique, and AdaBoost). We conduct experiments using 15 data sets from various application domains, four base learners, and four evaluation metrics. RUSBoost and SMOTEBoost both outperform the other procedures, and RUSBoost performs comparably to (and often better than) SMOTEBoost while being a simpler and faster technique. Given these experimental results, we highly recommend RUSBoost as an attractive alternative for improving the classification performance of learners built using imbalanced data.
Constructing classification models using skewed training data can be a challenging task. We present RUSBoost, a new algorithm for alleviating the problem of class imbalance. RUSBoost combines data ...sampling and boosting, providing a simple and efficient method for improving classification performance when training data is imbalanced. In addition to performing favorably when compared to SMOTEBoost (another hybrid sampling/boosting algorithm), RUSBoost is computationally less expensive than SMOTEBoost and results in significantly shorter model training times. This combination of simplicity, speed and performance makes RUSBoost an excellent technique for learning from imbalanced data.
Measurements of the production of forward jets from transversely polarized proton collisions at s=500 GeV conducted at the Relativistic Heavy Ion Collider (RHIC) are reported. Our measured jet cross ...section is consistent with hard scattering expectations. Our measured analyzing power for forward jet production is small and positive, and provides constraints on the Sivers functions that are related to partonic orbital angular momentum through theoretical models.
The importance of measuring outcomes after injury beyond mortality and morbidity is increasingly recognized, though underreported in humanitarian settings. To address shortcomings of existing outcome ...measures in humanitarian settings, the Activity Independence Measure-Trauma (AIM-T) was developed, and is structured in three subscales (i.e., core, lower limb, and upper limb). This study aimed to assess the AIM-T construct validity (structural validity and hypothesis testing) and reliability (internal consistency, inter-rater reliability and measurement error) in four humanitarian settings (Burundi, Iraq, Cameroon and Central African Republic). Patients with acute injury (n = 195) were assessed using the AIM-T, the Barthel Index (BI), and two pain scores. Structural validity was assessed through confirmatory factor analysis. Hypotheses were tested regarding correlations with BI and pain scores using Pearson correlation coefficient (PCC) and differences in AIM-T scores between patients' subgroups, using standardized effect size Cohen's d (d). Internal consistency was assessed with Cronbach's alpha (α). AIM-T was reassessed by a second rater in 77 participants to test inter-rater reliability using intraclass correlation coefficient (ICC). The results showed that the AIM-T structure in three subscales had an acceptable fit. The AIM-T showed an inverse weak to moderate correlation with both pain scores (PCC<0.7, p≤0.05), positive strong correlation with BI (PCC≥0.7, p≤0.05), and differed between all subgroups (d≥0.5, p≤0.05). The inter-rater reliability in the (sub)scales was good to excellent (ICC 0.86-0.91) and the three subscales' internal consistency was adequate (α≥0.7). In conclusion, this study supports the AIM-T validity in measuring independence in mobility activities and its reliability in humanitarian settings, as well as it informs on its interpretability. Thus, the AIM-T could be a valuable measure to assess outcomes after injury in humanitarian settings.
We present a quantitative assessment of the impact a future electron-ion collider would have in the determination of parton distribution functions in the proton and parton-to-hadron fragmentation ...functions through semi-inclusive deep-inelastic electron-proton scattering data. Specifically, we estimate the kinematic regions for which the forthcoming data are expected to have the most significant impact in the precision of these distributions, computing the respective correlation and sensitivity coefficients. Using a reweighting technique for the sets of simulated data with their realistic uncertainties for two different center-of-mass energies, we analyze the resulting new sets of parton distribution functions and fragmentation functions, which have significantly reduced uncertainties.