IEEE Symposium on Explainable Data Analytics in Computational Intelligence (EDACI)

The latest advances in Machine Learning are reaching critical areas such as medicine, criminal justice systems, financial markets and many other real applications. From a mathematical point of view, the search of the model focuses on the minimization of a cost function or the maximization of a likelihood function. Thus, the performance of the model is measured almost exclusively on the results we can get according to some rightly chosen metrics. This tendency has led to more and more sophisticated algorithms to the cost of explainability (Interpretability). Having an accurate model is good, but explanations lead to better products. There is an increasing concern about the acceptance of the computation intelligent learning models, where the explainability is the key measures for evaluating models.
In recent years, the advancements in computational intelligence have allowed researchers to tackle data driven problems with explainability and integrate efficient optimization algorithms for solving them. Due to the long-term memory, nonlocality, and weak singularity fractional differential operator, there is an increasing applications on fractional calculus based computation intelligent models. It is also an interesting research area to pay more attention on the interpretability aspect. From Explainable Data Analytics concern, this symposium aims to highlight the latest results from world leading research labs, academia and industry in the fields of Computational Intelligence, whose issues include corresponding efficient neural network methods, evolutionary algorithms and Neuro-fuzzy optimization techniques. In order to encourage research interactions, we welcome submissions describing innovative operations research methods that are able to provide state-of-the-art solutions to the above mentioned issues as well. Researches incorporating real-world applications are also highly encouraged.

Topics

The symposium will cover all the issues, researches and developments of the state-of-the-art EDACI-based learning models in solving various problems. CI application areas include, but are not limited to:

• Neural Network Learning Models
• Feature Analysis based Neural Networks Models
• Dimensionality reduction and analysis of large and complex data
• Fractional Evolutionary Optimization Computation
• Feature Analysis based Evolutionary Computation Algorithms
• Feature Analysis based Neuro-fuzzy Systems
• Extracting Understanding from Large-scale Data Resources
• Feature learning and feature engineering
• Time Series and System Modeling
• Flexible Neuro-fuzzy Systems
• Interpretability of Fuzzy Rule-based Systems for Nonlinear Modeling
• Dimensionality Reduction and Analysis of Large and Complex Dataset
• Neural Networks, Fuzzy and Evolutionary based Explainable Control Systems
• Optimization of big data in complex systems
• Expert and Decision Support Systems
• System Identification and Learning