The experimental results show that, the method is effective in clustering while dealing with undefinedboundary problems, and is powerful in avoiding noise.
实验结果表明，该算法对聚类边界不清晰的数据集可获得较精确的聚类划分，同时具有很强的噪声抑制能力。
2
In addition, by using the hole edges conditions, the relationship between undefined coefficients is set up, to reduce the number of those coefficients, and only the boundary collocation is needed.
同时，利用内孔边界条件，建立待定系数间的关系式，以减少待定系数的数目，并且只需要在外边界配点。
3
Artificial immune network clustering is often ineffective when there is noise or undefined cluster boundary in the data.