Affiliation:
1. Lovely Professional University, India
2. Acharya University, Uzbekistan
Abstract
E-commerce applications are widely used in shopping, money transfers, and other purposes. These applications store the user data to provide good results and recommend new products to the user. The user data saved on the server are critical and must provide proper security, safety, and availability. Any attack on these data may damage users' privacy. Machine learning can analyze and detect any attack on these data and applications. A machine learning model uses high-impact features to make a good model and effective prediction. Selecting the most critical data attributes is known as feature selection. Effective feature selection is a vital component in the process of developing a model with a high level of accuracy. This chapter explains the feature selection process and compares different feature selection techniques. This chapter analyzes filter-matched, wrapper-matched, and embedded methods and their variations. This chapter discussed the essential features concerning the security of e-commerce applications and e-commerce data.
Reference32 articles.
1. Al-Tashi, Q., Md Rais, H., Abdulkadir, S. J., Mirjalili, S., & Alhussian, H. (2020). A review of grey wolf optimizer-based feature selection methods for classification. Evolutionary Machine Learning Techniques: Algorithms and Applications, 273–286.
2. A two-layer feature selection method using Genetic Algorithm and Elastic Net
3. A novel key distribution scheme against storage-bounded adversaries using attack probabilities.;M. F.Bajestani;Turkish Journal of Electrical Engineering and Computer Sciences,2016
4. A systematic review of genetic algorithm-based multi-objective optimisation for building retrofitting strategies towards energy efficiency
5. Fused lasso for feature selection using structural information