A Weighted Bacterial Colony Optimization for Feature Selection

Feature selection is essentially important for high dimensional feature characterization problems. In this paper, we propose a weighted feature selection algorithm based on bacterial colony optimization (BCO) for dimensionality reduction. The weighted strategy is used for reducing the redundant features as well as increasing the classification performance, which considers the frequency of features being selected by bacterial colony optimization(BCO) as well as the repeated appearance in the same individual. The contributions of features in classification will be evaluated and kept in ‘Achieve’. The learning mechanism used in BCO considers the randomness which avoids the ignorance of unseen features as well as disengages from the local optimal error. Benchmark datasets with varying dimensionality are selected to test the effectiveness of the proposed feature selection method. The significance of the proposed weight feature selection algorithm is verified by comparing with three recently proposed population based feature selection algorithms.

[1]  Shichao Zhang,et al.  AI 2005: Advances in Artificial Intelligence, 18th Australian Joint Conference on Artificial Intelligence, Sydney, Australia, December 5-9, 2005, Proceedings , 2005, Australian Conference on Artificial Intelligence.

[2]  Anil K. Jain,et al.  Dimensionality reduction using genetic algorithms , 2000, IEEE Trans. Evol. Comput..

[3]  Xin Jin,et al.  Wavelet-based feature extraction using probabilistic finite state automata for pattern classification , 2011, Pattern Recognit..

[4]  Jo Røislien,et al.  Feature extraction across individual time series observations with spikes using wavelet principal component analysis , 2013, Statistics in medicine.

[5]  Mengjie Zhang,et al.  Particle swarm optimisation for feature selection in classification: Novel initialisation and updating mechanisms , 2014, Appl. Soft Comput..

[6]  Mansour Sheikhan,et al.  Time series prediction using PSO-optimized neural network and hybrid feature selection algorithm for IEEE load data , 2012, Neural Computing and Applications.

[7]  Adel Al-Jumaily,et al.  Feature subset selection using differential evolution and a statistical repair mechanism , 2011, Expert Syst. Appl..

[8]  Sreeram Ramakrishnan,et al.  A hybrid approach for feature subset selection using neural networks and ant colony optimization , 2007, Expert Syst. Appl..

[9]  Hong Hu,et al.  Ant Colony Optimization Combining with Mutual Information for Feature Selection in Support Vector Machines , 2005, Australian Conference on Artificial Intelligence.

[10]  Kevin M. Passino,et al.  Biomimicry of bacterial foraging for distributed optimization and control , 2002 .

[11]  Fuhui Long,et al.  Feature selection based on mutual information criteria of max-dependency, max-relevance, and min-redundancy , 2003, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Kevin Tickle,et al.  Solving the traveling salesman problem using cooperative genetic ant systems , 2012, Expert Syst. Appl..

[13]  Erik Schaffernicht,et al.  Weighted Mutual Information for Feature Selection , 2011, ICANN.

[14]  Ben Niu,et al.  An Adaptive Bacterial Foraging Optimization Algorithm with Lifecycle and Social Learning , 2012 .

[15]  Yishi Zhang,et al.  Feature subset selection with cumulate conditional mutual information minimization , 2012, Expert Syst. Appl..

[16]  Roberto Battiti,et al.  Using mutual information for selecting features in supervised neural net learning , 1994, IEEE Trans. Neural Networks.

[17]  Hong Wang,et al.  Bacterial Colony Optimization , 2012 .

[18]  George C. Runger,et al.  A time series forest for classification and feature extraction , 2013, Inf. Sci..

[19]  Kin Keung Lai,et al.  Bankruptcy prediction using SVM models with a new approach to combine features selection and parameter optimisation , 2014, Int. J. Syst. Sci..

[20]  Petros Koumoutsakos,et al.  Optimization based on bacterial chemotaxis , 2002, IEEE Trans. Evol. Comput..