IEEE Computational Intelligence Society

IEEE CIS Task Force on

Evolutionary Computation for Feature Selection and Construction

in Evolutionary Computation Technical Committee , IEEE Computational Intelligence Society



In machine learning and data mining, the quality of the input data determines the quality of the output (e.g. accuracy), known as the GIGO (Garbage In, Garbage Out) principle. For a given problem, the input data of a learning algorithm is almost always expressed by a number of features (attributes or variables). Therefore, the quality of the feature space is a key for success of any machine learning and data algorithm.

Many real-world problems involve a large number of features/variables, which leads to the problem known as "the curse of dimensionality". However, not all features are essential since many of them are redundant or irrelevant, and the useful features are typically not equally important. This problem can be solved by feature selection to select a small subset of original features, or feature construction to construct a smaller set of high-level features using the original low-level features and mathematical or logical operators. Feature selection and construction are challenging tasks because of the large search space and feature interaction problems. Due to the powerful search abilities and/or flexible solution encoding/representation schemes, there has been increasing interest in using evolutionary computation (EC) techniques to solve feature selection and construction problems. However, the dimensionality and the complexity of the data in real-world problems grows fast in recent years, which requires novel effective and efficient approaches to addressing new challenges in this area.



Feature selection and construction are important tasks in many areas, such as Data Mining, Machine Learning, Image Processing and Analysis, Statistics, Operation Research, Biology, Engineering, Finance, and Business. Researchers from these areas have started investigating EC techniques to solve feature selection and construction problems, but these researchers attend different events and activities. This task force would be an outstanding platform for them to share knowledge, exchange ideas, transfer tools, and generate new research lines.

The objectives of this task force are:


Anticipated interests

The theme of this task force is EC for feature selection and construction, covering all different EC paradigms. Topics of interest include but are not limited to:



Coming Events:

Past Events:



Bing Xue

Victoria University of Wellington, New Zealand,

Vice Chairs

Yaochu Jin

University of Surrey, United Kingdom,

Mengjie Zhang

Victoria University of Wellington, New Zealand,



Kay Chen Tan, National University of Singapore, Singapore

Yew-Soon Ong, Nanyang Technological University, Singapore

Hisao Ishibuchi, Tohoku University, Japan

Carlos A. Coello Coello, Cinvestav-IPN, Mexico

Xiaodong Li, RMIT University, Australia

Krzysztof Krawiec, Poznan University of Technology, Poland

Brijesh Verma, Central Queensland University, Australia

Stefano Cagnoni, Universita degli Studi di Parma, Italy

Stewart W. Wilson, Prediction Dynamics, USA

Zexuan Zhu, Shenzhen University, China

Kai Qin, RMIT University, Australia

Kourosh Neshatian, University of Canterbury, New Zealand

Andy Song, RMIT University, Australia

Ivy Liu, Victoria University of Wellington, New Zealand

Lin Shang, Nanjing University, China

Yi Mei, Victoria University of Wellington, New Zealand

Zhongyi Hu, Wuhan University, China

Emrah Hancer, Department of Computer Engineering, Erciyes University, Kayseri 38039, Turkey

Harith Al-Sahaf, Victoria University of Wellington, New Zealand

Qi Chen, Victoria University of Wellington, New Zealand

Urvesh Bhowan, IBM, Ireland

Liam Cervante, Google, United Kingdom