Xiaoguang (Benjamin) Wang
I work at UYUN. I did my Ph.D. in Computer Science at University of Ottawa under Professor Stan Matwin. I was co-advised by Professor Nathalie Japkowicz. My research interests include machine learning and applications of AI, as well as other use cases.
Projects
I used to work at Alibaba DAMO academy as an AI research scientist. After that I worked as the chief scientist in UYUN. Since 2015 me and my team has implemented dozens of AI projcets, including AIOps, Zhima credit, Intelligent city, and so on.
Research
My research focus on imbalanced learning and multiple instance learning. Here are githubs about these two topics：
Publications
For a complete list, see Google Scholar.Articles published or accepted in refereed Journals
Automatic Target Recognition Using Multiple-Aspect Sonar Images. Journal of Artificial Intelligence and Soft Computing Research. 12 pages. (2014). Wang, X., Liu, X., Japkowicz, N., & Matwin, S.
Boosting support vector machines for imbalanced data sets. Knowl. Inf. Syst. 25(1): 1-20 (2010). Wang, X., Japkowicz, N.
Meta-MapReduce for Scalable Data Mining. Journal of Big Data. 12 pages. (2015) Wang, X., Japkowicz, N. Liu, X., Wang, X., Matwin, S., & Japkowicz, N.
Papers in refereed conference proceedings
A Multi-View Two-level Classification Method for Generalized Multi-instance Problems.. Wang, X., Liu, X., Matwin, S. & Japkowicz, N., Guo, H. 2014 IEEE International Conference on Big Data, 104-111.
Vessel Route Anomaly Detection with Hadoop MapReduce.. Wang, X., Liu, X., Bo Liu, Erico N. de Souza & Matwin, S. 2014 IEEE International Conference on Big Data, 25-30.
Applying Instance-weighted Support Vector Machines to Class Imbalanced Datasets.. Wang, X., Liu, X., Matwin, S. & Japkowicz, N. 2014 IEEE International Conference on Big Data, 112-118.
A Distributed Instance-weighted SVM Algorithm on Large-scale Imbalanced Datasets.. Wang, X., Liu, X. & Matwin, S. 2014 IEEE International Conference on Big Data, 45-51.
Automatic Target Recognition using multiple-aspect sonar images.. Wang, X., Liu, X., Japkowicz, N., Matwin, S. & Nguyen B. IEEE Congress on Evolutionary Computation 2014: 2330-2337.
Resampling and Cost-Sensitive Methods for Imbalanced Multi-instance Learning.. Wang, X., Liu, X., Japkowicz, N., & Matwin, S. 2013 IEEE International Conference on Data Mining (ICDM), 9 pages.
Ensemble of Multiple Kernel SVM Classifiers.. Wang, X., Liu, X., Japkowicz, N., & Matwin, S. Canadian Conference on AI 2014: 239-250.
Cost-Sensitive Boosting Algorithms for Imbalanced Multi-instance Datasets.. Wang, X., Matwin, S., Japkowicz, N., & Liu, X. In Advances in Artificial Intelligence (pp. 174-186). Springer Berlin Heidelberg.
Meta-learning for Large Scale Machine Learning with MapReduce.. Wang, X., Matwin, S., Japkowicz, N., & Liu, X. 2013 IEEE International Conference on Big Data, 6 pages.
An Ensemble Method Based on AdaBoost and Meta-Learning.. Liu, X., Wang, X., Japkowicz, N., & Matwin, S. In Advances in Artificial Intelligence (pp.278-285). Springer Berlin Heidelberg.
Using SVM with Adaptively Asymmetric Misclassification Costs for Mine-Like Objects Detection.. Wang, X., Shao, H., Japkowicz, N., Matwin, S., Liu, X., Bourque, A., & Nguyen, B. (2012). In Machine Learning and Applications (ICMLA), 2012 11th International Conference on (Vol. 2, pp. 78-82). IEEE.
Boosting Support Vector Machines for Imbalanced Data Sets. . Wang, X., Japkowicz, N. (2012). ISMIS 2008: 38-47 (The Best Paper Award).
Book Chapters
Automated Mine-like Objects Recognition Using Instance-weighted Boosting SVM on Imbalanced Multiple Instance Dataset. Wang, X., Liu, X., Japkowicz, N., & Matwin, S. (2015). Recent Advances in Computational Intelligence in Defense and Security. Submitted, 30 pages.
Patents
Training method, credit estimation method and the device of credit evaluation model. Patent number : CN107301577A
Model training method, sample balancing method, model training device, sample balancing device and personal credit scoring system Patent number : CN106909981B
Model data updating method, device and system Patent number : CN107229966B
Model training method, apparatus and system and sample set optimization method, device Patent number : CN106934413A
A kind of Risk Forecast Method and equipment Patent number : CN106779272A
Method and device for determining user intention based on user voice information Patent number : CN108205525B
The method and device that the belonging kinds of data are predicted Patent number : CN107203774A
The sorting technique and system of data Patent number : CN106934410A
A kind of method and device for optimizing user credit model modeling process Patent number : CN106997484A
A kind of Feature Selection method and device Patent number : CN107169571A
Feature engineering strategy determination method and device Patent number : CN107168965B
User characteristics sorting technique, user credit appraisal procedure and the device of user credit model Patent number : CN106997472A
A kind of method and device for screening user characteristics Patent number : CN106874286A
A kind of information extracting method and device Patent number : CN107133207A
User group classification method and device Patent number : CN106897282B
User credit model establishing method and device Patent number : CN107203916B
object grouping method, model training method and device Patent number : CN106874925A
Intelligent operation and maintenance system based on data middling platform technology Patent number : CN112182077B
Intelligent operation and maintenance framework system based on AIOps Patent number : CN112181960B
Three-dimensional microscopic road network generation method capable of realizing real-time interaction Patent number : CN111535099B
Method and system for adaptively calculating IT intelligent operation and maintenance health index Patent number : CN113360358A
Fault root cause positioning method and system based on multidimensional data map Patent number : CN113360722A
Semi-supervised man-machine combined operation and maintenance fault library generation method and system Patent number : CN112783865A
Disk capacity prediction method for identifying manual cleaning behavior based on second-order difference method Patent number : CN113157204B
