Improving the Accuracy of Feature Selection in Big Data Mining Using Accelerated Flower Pollination (AFP) Algorithm.

Area of Science:

• Computer Science
• Data Science
• Artificial Intelligence

Background:

• Big data analytics faces challenges with high-dimensional data, making traditional methods infeasible for real-time mining.
• Feature selection is crucial for reducing computational load and improving data mining efficiency.
• Optimal feature subset formulation in high-dimensional data leads to exponential growth and intractable computational demands.

Purpose of the Study:

• To propose a novel, lightweight feature selection mechanism for big data mining.
• To address the computational challenges associated with optimal feature selection in high-dimensional datasets.
• To improve the accuracy and reduce processing time in big data mining.

Main Methods:

• A novel lightweight mechanism for feature selection is employed.
• The accelerated flower pollination (AFP) algorithm is utilized for feature selection in big data mining.
• The proposed method is evaluated on large, high-dimensional datasets.

Main Results:

• The accelerated flower pollination (AFP) algorithm demonstrates improved accuracy in feature selection.
• The proposed method significantly reduces processing time compared to traditional approaches.
• Effective performance is observed even with large datasets characterized by high dimensionality.

Conclusions:

• The accelerated flower pollination (AFP) algorithm offers an efficient solution for feature selection in big data analytics.
• This approach effectively mitigates the computational burden of mining high-dimensional data.
• The method enhances the feasibility of real-time data mining applications.