Closed snvv closed 1 month ago
If you have a supervised dataset with texts with positive/negative connotations, you can build a classification model with this R package and use the embeddings to find similarity of these to being positive/negative. If you don't have supervised dataset, you can use this R package to build wordvectors and find similarities to a small set of words which are positive/negative to construct your dictionary.
Hello It would be very helpful to provide an example for sentiment scoring. I have to create a dictionary for sentiment analysis using txt_sentiment function. Thank you