|
I. Aim and Scope
Today, big data has become capital in both academia and industry fields, which is changing our world and the way we live at an unprecedented rate. Recent advances in computing technology allow us to gather and store large amounts of information from various fields, such as Internet, sensor monitoring systems, social networks, mobile communication systems, and transportation systems. Since big data contains greater variety arriving in increasing volumes and with ever-higher velocity, it is essential to develop new data mining and knowledge discovery techniques, and especially using evolutionary computation techniques help in the information retrieval process in a better way compared to traditional retrieval techniques.
Considering big data mining technologies, it still meet some serious problems requiring to be tackled, such as classification, clustering, regression, associate rules mining, and frequent pattern mining. The ability of evolutionary algorithms includes managing a set of solutions, attending multiple objectives, as well as their ability to optimize various values, which allows them to fit very well some parts of the big data mining problems. The global search performed by evolutionary methods frequently provides a valuable complement to the local search of non-evolutionary methods, and combinations of data mining and evolutionary methods often show particular promise in practice. Evolutionary data mining emphasizes the utility of different evolutionary algorithms to various facets of data mining from databases, ranging from theoretical analysis to real-life applications. The primary motivation of applying evolutionary data mining procedures is to automatically and effectively extract relevant information from a voluminous pool of data, which can conduct global search in the solution space (for example, rules, or another form of knowledge representation). It has become a hot trend to address the classical and new-emerging data mining problems in big data with different evolutionary algorithms. The benefits of investigating the combination of data mining and evolutionary computation in the big data scenario have potential to apply in multiple research disciplines and industries, including transportation, communications, social networks, medical health, and so on.
II. Themes
The goal of this Special Issue aims to provide an unique opportunity to present the work on state-of-the-art of evolutionary data mining algorithms in the area of big data processing. This will provide a snapshot of the latest advances in the contribution of evolutionary frameworks in big data mining applications to solve optimization problems and optimize criteria. The selected papers will be beneficial to both academia and industry, for delivering the significant research outcomes and inspiring new real-world applications.
The topics of interest include, but are not limited to:
Evolutionary multi-modal optimizations for data mining in big data
Dimensionality reductions using evolutionary computation
Large-scale neuro-evolutionary algorithms for data mining in big data
Evolutionary methods for training deep networks with application to data mining tasks
Data-driven large scale optimizations for data mining in big data
Feature selection and extraction methodologies to attribute reductions in high-dimensional and large-scale data
Evolutionary multi/many-objective optimization for data mining in big data
Evolutionary constrained optimization for data mining
Associate rule mining using evolutionary computation
Convergence analysis of evolutionary algorithms for data mining
Parallelized and distributed realizations of evolutionary algorithms
Adaptive knowledge discoveries using advanced metaheuristic and evolutionary algorithms
Semi-supervised learning and transfer learning with evolutionary optimization
Evolutionary fuzzy systems for big data
Ensemble learning with evolutionary optimization
Quantum-inspired evolutionary algorithm for data mining
Co-evolutionary algorithm for data mining in big data
Multi-label, multi-instance and multi-view problems using the evolutionary algorithm
Low quality and/or noisy big data mining problems using the evolutionary algorithm
Evolutionary optimizations with the dynamic parameter adaptation based on fuzzy systems
Recommender system, graph mining, data stream mining and time series analysis with evolutionary algorithms
Real-world big data applications using evolutionary data mining approaches
III. Submissions
The manuscripts should be prepared according to the “Guide for Authors” section of the journal found at: https://www.elsevier.com/journals/swarm-and-evolutionary-computation/2210-6502/guide-for-authors/ and submission should be done through the journal’s submission website: https://www.evise.com/profile/#/SWEVO/login/ by selecting “VSI: Evolutionary Data Mining” and also clearly indicating the full title of this special issue “Evolutionary Data Mining for Big Data” in comments to the Editor-in-Chief.
Each submitted paper will be reviewed by expert reviewers. Submission of a paper will imply that it contains original unpublished work and is not being submitted for publication elsewhere.
IV. Important Dates
Initial Paper Submission: April 30, 2019
Initial Paper Decision: July 31, 2019
Revised Paper Submission: September 30, 2019
Final Paper Decision: November 30, 2019
Final Paper Submission: December 31, 2019
Publication Date (Provisional): Early 2020
V. Guest Editors
Weiping Ding
Nantong University, China
Email address: [email protected]
Gary G. Yen
Oklahoma State University, U.S.A.
Email address: [email protected]
Xinye Cai
Nanjing University of Aeronautics and Astronautics, China
Email address: [email protected]
Zehong Cao
University of Technology Sydney, Australia
Email address: [email protected]
|
