Weka - no class and no cluster assign -


I am working on a file to predict the treatment methods for patients diagnosed with diabetes (level) Is 1 to 10). There are 8 different treatment recommendations (256 possible outcomes) and I have to cluster them (I have 21 features from the original file.). So I used SimpleKMan with 1 with 1. The problem is that I have been given "no class" for some groups;

cluster window]! [Enter image details here

Also, when I classify it for evaluation, I have a similar problem, "no cluster" is not assigned to the class and I also lose some data. For example, there are 940 examples but I have 876 after classification .

classified window]! [Enter image details here

But the illusion matrix displays the exact number. I do not know that it is related, but it can help solve the question. I have used AddCluster method because all my properties are numeric and I need an extra column from the original file to display "treatment cluster" (22nd attributes). That's why I run simple Cayman and Cross-Validation with this new extra feature which is my class too.

Thank you very much for your help!

It seems that a class can only be applied to zero or one cluster, as a result, for example For class 9 is being applied to cluster 7, but not all sections of 9 classes in cluster 8 are being assigned because it is being allocated in another class. The simple Cayman model appears to specify the cluster that produces the minimum classification error on the supplied data.

This issue has been raised before, where one-to-many allocation to override the solution evaluation model

Comments

Popular posts from this blog

java - ImportError: No module named py4j.java_gateway -

python - Receiving "KeyError" after decoding json result from url -

.net - Creating a new Queue Manager and Queue in Websphere MQ (using C#) -