RETRACTED: An improved DBSCAN algorithm based on cell-like P systems with promoters and inhibitors

Area of Science:

• Computer Science
• Bio-inspired Computing
• Artificial Intelligence

Background:

• Density-based spatial clustering of applications with noise (DBSCAN) is effective for finding arbitrary-shaped clusters and removing noise.
• Membrane computing, a bio-inspired field, develops computational models called P systems from biological cells, offering parallel and distributed computing capabilities.

Purpose of the Study:

• To improve the efficiency and performance of the DBSCAN algorithm for large-scale cluster analysis.
• To integrate membrane computing principles, specifically parallel evolution and hierarchical structures, into DBSCAN.

Main Methods:

• The DBSCAN algorithm was enhanced using cell-like P systems featuring a hierarchical membrane structure.
• Promoters and inhibitors were employed within the P systems to regulate the parallel evolution of objects.
• A parallel evolution mechanism was integrated into the DBSCAN framework.

Main Results:

• The enhanced DBSCAN algorithm demonstrated strong performance in analyzing large datasets and big clusters.
• The time complexity was significantly improved from O(n^2) for conventional DBSCAN to O(n).
• The integration of membrane computing concepts led to a more efficient clustering solution.

Conclusions:

• The proposed enhanced DBSCAN algorithm effectively handles big cluster analysis with improved time complexity.
• Membrane computing models, particularly hierarchical frameworks and parallel evolution mechanisms, offer a promising approach for enhancing conventional algorithms.
• This research highlights the potential of bio-inspired computing for advancing data analysis techniques.