edda extension

guzin_kavrukkoc

I am trying to develop a model using EDDA topic Model . However, in the help tab, it says example file should contain term vector based on occurrences and a binomial label. I also says use process documents from file.

I have done all things except labeling, because set role operator can not be used after process documents from files operator. In fact in this operator I tried to extract sentiments and use it as a label but I could not convert it to a label.

When I run the first part of the model, an error message appeared saying that "use process document from files operator" which I have already used. I could not solve this issue.
Could you please help me about this problem or send me an example model in which the operator edda topic model for binomial classification was used?
or send me an idea to obtain topics from documents in rapid miner?

bhanus

I have a similar problem. I am using Process Document from FIles operator. I set Term Occurences for vector creation, and add meta information as instructed. I get an error message "Input Example Set corrupt for TM construction". Not exactly sure what is going on. Were you able to resolve your issue?