br The proposed research su ciently compared and promoted
The proposed research su ciently compared and promoted un-derstanding of typical online learning algorithms. Complicated pa-rameter tuning is one of the main concerns in improving online learning performances. The GAOGB algorithm addresses issues of parameter selection, conforms to certain convergence bounds and maintains su cient adaptiveness. The experimental results suggest that the GAOGB structure is effectively configured. In addition, this AUY922 (NVP-AUY922) research validates that parameter tuning is a critical step in online learning algorithms.
The future work is discussed from two aspects: the BC research and online boosting. Most current BC research topics focus on of-fline and supervised learning, requiring long retraining time and massive manual labeling work. Online learning models are recom-mended for more BC research topics, and label-free learning tech-niques show promise in saving human labor. For online boosting, enhancing the superiority of GAOGB over OGB is an interesting challenge, which may be tackled through adapting a heuristic al-gorithm for online parameter optimization in a more e cient way. From another perspective, configuring effective parameter free al-gorithms is also a promising direction.
Friedman, M. (1937). The use of ranks to avoid the assumption of normality im-plicit in the analysis of variance. Journal of the American Statistical Association, 32(200), 675–701.
Lavanya, D., & Rani, K. U. (2012). Ensemble decision tree classifier for breast can-cer data. International Journal of Information Technology Convergence and Services, 2(1), 17.
Parag, T., Porikli, F., & Elgammal, A. (2008). Boosting adaptive linear weak classi-fiers for online learning and tracking. In Computer vision and pattern recognition (CVPR), 2008 IEEE conference on (pp. 1–8). IEEE.
Pereira, D. C., Ramos, R. P., & Do Nascimento, M. Z. (2014). Segmentation and detec-tion of breast cancer in mammograms combining wavelet analysis and genetic algorithm. Computer Methods and Programs in Biomedicine, 114(1), 88–101.
Puuronen, S., Terziyan, V., & Tsymbal, A. (1999). A dynamic integration algorithm for an ensemble of classifiers. In International symposium on methodologies for intelligent systems (pp. 592–600). Springer.
Ravdin, P. M., & Clark, G. M. (1992). A practical application of neural network anal-ysis for predicting outcome of individual breast cancer patients. Breast Cancer Research and Treatment, 22(3), 285–293.
SEER (2017). Surveillance, epidemiology, and end results (SEER) program research data (1973–2014). www.seer.cancer.gov. DCCPS, Surveillance Research Program, Surveillance Systems Branch, released April 2017, based on the November 2016 submission.
United NationsDepartment of Economic and Social Affairs, Population Divi-sion (2017). World population prospects: The 2017 revision, key findings and advance tables. Working Paper ESA/P/WP/248.
West, D., Mangiameli, P., Rampal, R., & West, V. (2005). Ensemble strategies for a medical diagnostic decision support system: A breast cancer diagnosis applica-tion. European Journal of Operational Research, 162(2), 532–551.
machine learning repository. https://archive.ics.uci.edu/ml/datasets/ BreastCancerWisconsin(Diagnostic).
Contents lists available at ScienceDirect
journal homepage: www.elsevier.com/locate/canlet
A feedforward relationship between active Rac1 and phosphorylated Bcl-2 is T critical for sustaining Bcl-2 phosphorylation and promoting cancer progression
Stephen Jun Fei Chonga,b, Jolin Xiao Hui Laia, Jianhua Qua,b, Jayshree Hirparad, Jia Kanga,c, Kunchithapadam Swaminathane, Thomas Lohf, Ansu Kumarg, Shireen Valih, Taher Abbasih, Shazib Pervaiza,b,c,i,∗ a Department of Physiology, Yong Loo Lin School of Medicine, National University of Singapore (NUS), Singapore b Medical Science Cluster Cancer Program, Yong Loo Lin School of Medicine, National University of Singapore (NUS), Singapore