Process modeling and prediction presents a crucial issue to develop adaptive strategies in coping with industrial manufacturing and production lines. However, complex processes in industry are often hard to model using conventional mathematical techniques and algorithms on their own. This project aims to exploit a hybrid approach using learning techniques based on computational intelligence to build knowledge-based models and associated reasoning mechanisms for process modeling, prediction and classification. The key techniques employed in the research will include: fuzzy computing, case-based reasoning, nature-inspired optimization, and perhaps also probabilistic inference to accommodate stochastic property of processes.
A new case-based reasoning method based on dissimilar relations (Jul 2014) Ning Xiong WSEAS Transactions on Systems (WT-SYS-13)
Fuzzy relational learning: a new approach to case-based reasoning (Jul 2013) Ning Xiong, Liangjun Ma, Shouchuan Zhang The 10th International Conference on Fuzzy Systems and Knowledge Discovery (FSKD 2013)
Fuzzy Rule-Based Similarity Model Enables Learning from Small Case Bases (Jun 2013) Ning Xiong Applied Soft Computing, in press
Editorial: Special Issue on "Recent Advances in Intelligent Techniques" (Mar 2013) Yongmin Li , Ning Xiong, Haiying Wang , Lipo Wang International Journal of Intelligent Systems (IJIS-28-3)
Fuzzy dissimilarity learning in case-based reasoning (Dec 2012) Ning Xiong Proc. 3rd European Conference on Systems
Case-based reasoning and its relation to information fusion (Dec 2012) Ning Xiong Proc. International Conference on Computational Intelligence and Software Engineering