Altair RISE

A program to recognize and reward our most engaged community members

Nominate Yourself Now!

How do I classify an already classified data in Rapidminer

Hi everybody!

I made a text classification with Rapidminer. Now I need to categorize this classification again. How can I do that? Any idea?

For example: High school students are playing football.

The text is firstly classified as high "high school"

Now I need to classify it again as "football" in the same model like...

Students in high school and playin football

High school students are playin football = high school > football

Is it possible?

Thank you...

Find more posts tagged with

AI Studio

Getting Started

Accepted answers

All comments

Thomas_Ott

Hi Baran,

I had to move your post to the Getting Started Forum. Building Blocks is for people to share their RapidMiner Building Blocks.

When you said you made a text classification in RapidMiner, was that from a data set that you label by yourself and then loaded into RapidMiner, or was their a model like Naive Bayes that autoclassified the text for you?

baran

Thank you Thomas for your concern...

I load a training data and labeled myself... then used naive bayes to teach the machine. After that I load a test data, applied model to test data and got whether the model autoclassified the test data right or not. The result was perfect! Now I want to apply results of test data (the text) to a new classifier (with a new training data).

For example.

Text Category I Category II

High school students playing football High school Football

College students playing football College Football

University students playing basketball University Basketball

I have two training data one is for category I and second is for category II.

Can I see them in the same performance operator. Each column is saperate.

Thanks again

Thomas_Ott

Hmm, having two sets of labels in two columns might not work.

Can you merge them together into one column, like "highschool_football" and "College_football?" You could do that with a Generate Attributes operator, just create a new label column and do something like [attribute1]_[attribute2] in the expressoin field.