A Hybrid Algorithm for Feature Selection and Classification

<p>With a recent spread of intelligent information systems, massive data collections with lot repeated and unintentional, unwanted interference oriented are gathered huge feature set being operated. Higher dimensional inputs, on the other hand, contain more correlated variables, which might have negative impact model performance. In our Hybrid method selecting was developed by combining Binary Gravitational Search Particle Swarm Optimization (HBGSPSO) an Enhanced Convolution Neural Network Bidirectional Long Short Term Memory (ECNN-BiLSTM). proposed system, (BiLSTM) is introduced extracts hidden dynamic utilizes memory cells to think long-term historical after convolution process. this paper, thirteen well-defined datasets used from machine learning database UC Irvine evaluate efficiency system. The experiments conducted using K Nearest Neighbor (KNN) Decision Tree (DT) as classifiers outcome selected features. outcomes contrasted compared bio-enlivened calculations like Genetic Algorithm (GA), Grey Wolf Optimizer (GWO), protocol (PSO).</p> <p> </p>