I’ve been looking at the lexicon based sentiment analysis workflow and have some application questions.
Firstly, I want to perform a sentiment analysis of positive and negative terms as is completed in the lexicon based sentiment analysis workflow.
I want to essentially do the same thing as in the example but on my own documents. I can’t investigate the configurations on many of these nodes so I’m having difficulty adapting the workflow to my own problem.
Is there any more material I can read to guide me through performing a similar lexicon based sentiment analysis to the example workflow? If not, how do you suggest constructing a similar workflow?
Additionally, I need to search within documents for specific terms that I have stored in an excel file. I want to count the number of times any word from each category occurs and store this value in a table so that I can put it into an excel file.
Should I use the dictionary tagger to accomplish this?
I answered your question on the other topic, linked below. But I also wanted to point out here that if you’re looking for additional resources on how to do text processing and analytics with KNIME, we have a book dedicated to this topic that’s really good.
It’s available in our KNIME Press store, and is called From Words to Wisdom. I’ve gone through the entire book myself and it’s excellent - I’m not just saying that as a KNIME employee!
It should help you get oriented not only with the functions of individual nodes, but also approaches to particular text analysis strategies, whether it’s sentiment analysis, topic extraction, visualization, and even some word embedding.