Results of

Hello,

I am using "Apply Model (Documents)" object for getting sentiment score on tokenized data using my own created dictionary.
The output shows the tokenized data and the sentiment results, which is good, but no other attributes (that were entered to the input port) are presented.
This means that I cannot connect the sentiment results to the source.
How can I do that ?

Thanks,
DataMiner.

Find more posts tagged with

AI Studio

Text Mining + NLP

Sentiment Analysis

Accepted answers

MartinLiebig

Hi @DataMiner ,

i've fixed this in the internal build. It is currently under review. At the moment you would need to use Merge Attributes to "join" row based.

Best,

Martin

All comments

MarlaBot

Hi @DataMiner - this is MarlaBot. I found these great videos on our RapidMiner Academy that you may find helpful:

Instructional Video: Text Association Rules (Viewing time: ~10m)

Instructional Video: Loading Text into RapidMiner (Viewing time: ~6m)

Instructional Video: Document Similarity and Clustering (Viewing time: ~10m)

Instructional Video: Applying a Model to categorize Documents (Viewing time: ~11m)

Instructional Video: The RapidMiner MarketPlace - Extensions (Viewing time: ~5m)

Please LIKE my comment if it helps! 👇

MarlaBot

Papad

Hi @DataMiner
What to you mean to the source?
If you want to save them, you can do it easily using a "Write" operator to add it into your file or use "Store" in order to save it into a new repository.
Hope it helps,
Jim.

DataMiner

Hi Jim,

The input into the "doc" port of the "Apply Model (Documents)" is:
1. Text attribute of tokenized words
2. Customer ID, so I will be able to match between the sentiment score I get and the customer who generated the text.
The problem is that the output doesn't contain the Customer ID attribute.

Thanks,
DataMiner.

MartinLiebig

Hi @DataMiner ,

i've fixed this in the internal build. It is currently under review. At the moment you would need to use Merge Attributes to "join" row based.

Best,

Martin