Classify data without training data
显示 更早的评论
I have a data like
7.067584393
7.058334315
7.053180933
7.029897855
7.019380453
6.995541483
6.970042991
6.946210158
6.935758126
6.96556777
6.996278756
7.016032692
7.03634689
7.056638911
7.059383196
7.033728869
It is in excel file.. I would like to classify this data...if svm is possible then it is good..
回答（2 个）
Walter Roberson
2014-4-29
0 个投票
It is not possible. You could possibly cluster the data, but not classify it.
In order to classify data you need to have an algorithm that knows what to look for, or you need to create a mathematical model of what kinds of samples belong in which class. In order to create that kind of model, you need some sample data and you need to be told what the correct classification is for that sample data. The two together are used to estimate the parameters of the model. Then the parameters estimated can be applied to data for which the classification is not known, in order to find the most plausible class that the data belongs to.
类别
在 帮助中心 和 File Exchange 中查找有关 Statistics and Machine Learning Toolbox 的更多信息
产品
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!
Translated by 
