研究目的
To propose an improved spectral clustering algorithm based on Dynamic Tissue-like P System (ISC-DTP) that leverages the advantages of maximal parallelism in tissue-like membrane systems for more effective and robust clustering of high-dimensional data.
研究成果
The ISC-DTP algorithm demonstrates superior effectiveness and robustness compared to traditional K-means and spectral clustering algorithms, especially in handling non-convex data sets. The use of tissue-like membrane systems for determining the number of clusters and the improved K-means algorithm for initial cluster center selection contribute to its enhanced performance.
研究不足
The algorithm's performance is dependent on the initial selection of cluster centers and may still be affected by outliers and local optima, despite improvements. The computational efficiency, while better than traditional spectral clustering, is still slower than K-means for some datasets.
