关于聚类方法的问题
最近我开始研究数据挖掘中的聚类，并且研究了顺序聚类、层次聚类和 k 均值。
我还读到了一种将 k-means 与其他两种聚类技术区分开来的说法，说 k-means 不太擅长处理名义属性，但文本没有解释这一点。到目前为止，我唯一的区别是可以看到，对于 K 均值，我们会提前知道我们将需要恰好 K 个聚类，而我们不知道其他两种聚类方法需要多少个聚类。
那么有人能给我一些关于为什么存在这样的陈述的想法吗，即，k-means 在处理标称属性的示例时存在这个问题，有没有办法克服这个问题？
提前致谢。
recently I came to study clustering in data-mining and I've studied sequential clustering and hierarchical clustering and k-means.
I also read about a statement that distinguishes k-means from the other two clustering technique,saying k-means is not very good at dealing with nominal attributes,but the text didn't explain this point.So far,the only difference that I can see is that for K-means,we will know in advance we will need exactly K clusters while we don't know how many clusters we need for other two clustering methods.
So could anybody give me some idea here on why such statement exists,i.e.,k-means has this problem when dealing with examples of nominal attributes and is there a way to overcome this?
Thanks in advance.
如果你对这篇内容有疑问，欢迎到本站社区发帖提问 参与讨论，获取更多帮助，或者扫码二维码加入 Web 技术交流群。
绑定邮箱获取回复消息
由于您还没有绑定你的真实邮箱，如果其他用户或者作者回复了您的评论，将不能在第一时间通知您！
发布评论
评论（1）
k-means 算法通过取簇中所有点的平均值来计算簇质心。如果参数是名义参数，则不能取平均值。
有时，可以将标称值放入某种顺序中，然后映射到实际值。例如，一周中的几天可以映射到范围 [1.0 - 7.0]，但有时这是不可能的，例如具有值 [Windows、Linux、OSX] 的属性。
The k-means algorithm calculates cluster centroids by taking the mean values of all the points in the cluster. If a parameter is nominal then you can't take an mean value.
Sometimes nominal values can be put into a kind of order and then mapped to real values. For example, days of the week could be mapped onto the range [1.0 - 7.0], but then again sometimes that isn't possible, for example an attribute with values [Windows, Linux, OSX].