An improved clustering method for large-scale data based on artificial immune system

Data clustering plays an important role in machine learning, which is widely applied to pattern recognition, spatial data analysis, image processing, economics, document categorization, web mining, etc. However, it is still a big challenge for any clustering method to process high-dimensional and massive data. As a type of promising theory, artificial immune system (also called AIS) gradually becomes the research hotspot in related application fields. Its powerful information processing capacity often brings us new inspirations to solve real-world problems. For clustering analysis, AIS is employed to decrease the time complexity of handling original data, which results in simplifying the process of data processing and enhancing the computational efficiency. In this paper, we propose an improved approach to data clustering based on AIS. The efficiency evaluation is established for different datasets and parameters. The results show that the approach presented in this paper outperforms the classic aiNet.