Search

This paper investigates a new bacterial colony-based feature selection algorithm to improve the classification accuracy of customers for personalized products recommendation. An attribute learning strategy is developed in this study to update the feature related population. Specifically, the features can be weighted according to their historic contributions to both the individual- and group-based subsets. Additionally, the frequency of appearance is also recorded for the feature candidates to improve the diversity of feature distribution and avoid the over-fitting. Based on the weight-based feature indexes and frequency of appearance records, the performance of feature subsets are enhanced by replacing the features being repeatedly appeared in a same vector. To explore the feasibility of the proposed method for the missing feature problems, the objective of the optimization is to minimize the classification error using the acceptable number of features. K-Nearest Neighbor is employed as the learning technique to cooperate with the proposed feature selection method. The effectiveness of the proposed feature selection method is demonstrated by performing test on the datasets from UCI machine learning repository and real-world data from Amazon customer reviews of products. Compared with other seven feature selections methods, the proposed feature selection algorithm outperforms the other algorithms by achieving higher classification accuracy rate using smaller features. [ABSTRACT FROM AUTHOR]

Feature selection (FS) is an essential data-processing technique to reduce the number of features and improve the classification performance, but it is also a challenging problem because of the large search space and complex interactions between features. Bacterial based algorithms (BAs) are effective population based techniques known for their global searching capability. This paper proposes a novel bacterial algorithm based on control mechanisms and modified population updating strategies for feature selection in classification. The proposed new method, abbreviated as BAFS, employs three parameters to control the randomness of the population and reduce the computational complexity by avoiding the redundant searching for the optimal. To make the solutions suitable for feature selection, the strategies of reproduction and elimination are modified according to the classification performance and occurrence of features, respectively. Feature distribution is measured by the probability that features are appeared in the most promising subsets. The proposed bacterial based feature selection algorithm is used for selecting the best feature subsets on datasets with varying dimensionality. Comparison studies on five bacterial based algorithms indicate that the proposed BAFS outperforms other algorithms by achieving higher classification performance. [ABSTRACT FROM AUTHOR]