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.
Coulomb excitation of 74,76Zn Illana, A.; Zielińska, M.; Huyse, M. ...
Physical review. C,
10/2023, Letnik:
108, Številka:
4
Journal Article
Recenzirano
Odprti dostop
The first experiment using radioactive beams post-accelerated by the HIE-ISOLDE facility has enabled to obtain a precise set of B(E2) transition probabilities in neutron-rich 74,76Zn isotopes. The ...resulting B(E2; 2+1→0+1) values are consistent with those determined in earlier REX-ISOLDE measurements. While the B(E2; 4+1→2+1) transition probability in 76Zn is also in agreement with earlier Coulomb-excitation results, the value obtained for 74Zn is considerably lower. For the first time, a spectroscopic quadrupole moment of the 2+1 state was measured for an exotic nucleus in this mass region. A detailed comparison is presented with large-scale shell-model and Monte Carlo shell-model calculations.
The CERN/ISOLDE laser ion source Marsh, B. A.; Fedosseev, V. N.; Chrysalidis, K. ...
2017 Conference on Lasers and Electro-Optics (CLEO)
Conference Proceeding
Laser resonance photo-ionization an essential aspect of radioactive ion beam production for fundamental and applied physics research. The laser ion source of the ISOLDE facility, described here, is ...the most versatile of its type worldwide.
The neutron-rich isotopes ^{58-63}Cr were produced for the first time at the ISOLDE facility and their masses were measured with the ISOLTRAP spectrometer. The new values are up to 300 times more ...precise than those in the literature and indicate significantly different nuclear structure from the new mass-surface trend. A gradual onset of deformation is found in this proton and neutron midshell region, which is a gateway to the second island of inversion around N=40. In addition to comparisons with density-functional theory and large-scale shell-model calculations, we present predictions from the valence-space formulation of the ab initio in-medium similarity renormalization group, the first such results for open-shell chromium isotopes.
Bioinformatics and computational intelligence are undoubtedly remarkably fast growing fields of research and real-world applications with enormous potential for current and future developments. ...Bioinformatics Using Computational Intelligence Paradigms contains recent theoretical approaches and guiding applications of biologically inspired information processing systems (computational intelligence) against the background of bioinformatics. This carefully edited monograph combines the latest results of bioinformatics and computational intelligence, and offers promising cross-fertilization and interdisciplinary work between these growing fields.