Recent advances of statistics in computational intelligence (RASCI)
- PDF / 500,465 Bytes
- 3 Pages / 595.276 x 790.866 pts Page_size
- 78 Downloads / 203 Views
EDITORIAL
Recent advances of statistics in computational intelligence (RASCI) Muhammed J. A. Patwary1 · James N. K. Liu2 · Honghua Dai3
© Springer-Verlag GmbH Germany, part of Springer Nature 2017
Computational intelligence is usually considered to be the capability of a computer to learn a specific task from data or experimental observation. Computational intelligence plays a pivotal role in solving real-world complex problems. One of the main objectives of computational intelligence models is to discover the statistical disciplines hidden in the data sets. The most significant theoretical fundamental in developing different computational intelligence models is considered to be statistical analysis. There is a continuous scientific attempt to get better approaching into the cryptic information underneath the huge stack of statistical model and data that we come across to solve a specific problem. As a result, there has been a major change towards quantitative analysis of statistical methods as well as data through various computational approaches. Computational approaches that have been extremely popular and found significant application include Neural Network (NN) which provides with a representation framework for statistical construct, regression models that are based on pure statistical theory, Support Vector Machine (SVM) is based on framework of statistical learning, Bayesian method is also regarded as part of statistics, etc. On the whole, all the major tasks of computational intelligence can be represented as statistical methods and these methods can be explained from a statistical point of view. * Muhammed J. A. Patwary [email protected] James N. K. Liu [email protected] Honghua Dai [email protected] 1
Big Data Institute, College of Computer Science and Software Engineering, ShenZhen University, Shenzhen, China
2
Department of Computing, The Hong Kong Polytechnic University, Hong Kong, China
3
School of Information Technology, Deakin University, Melbourne Campus, 21 Burwood Highway, Burwood, VIC 3125, Australia
This special issue is an effort to bring together appealing works that use advanced statistical theory and methodology for handling the computational intelligence. After a strict blind peer review, we selected 12 papers for this issue. The following is a brief introduction to the 12 selected papers. The paper “Experimental study on generalization capability of extended naïve Bayesian classifier” by Si-si Chen, Juan-juan Cao, Li-li Gan, Qing-ge Song and Di Han, mainly investigates the Extended naïve Bayesian classifier (ENBC’s) generalization capability based on the real classification datasets. The analysis shows that ENBC is not stable and its aggregation weight is sensitive to the order of training samples and ENBC indeed has higher generalization capability than the existing NBC. The paper titled “NBWELM: naive Bayesian-based weighted extreme learning machine” authored by Jing Wang, Lin Zhang, Juan-juan Cao and Di Han discusses the generalization performance of imbalance
Data Loading...
