Apology Transformer #208
Open
Apology Transformer #208
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Description
Added a new apology transformer from INFO 4350 project with Prof. DNM. This transformer works to find an label apology types in a corpus at the utterance level.
It will identify and label these apologies in the utterance metadata.
Motivation and Context
This is a new transform that we created in order to study the presence and effects of apologies. When parsing through the Conversations Gone Awry Corpus, we noticed that quite a few utterances contained apologies, and then we noticed distinct patterns of their use. We were able to label five primary categories of apologies: Clarifying, Disagreement, Wrongdoing, Request, and Other (apologies that don't fit into the other four).
We were then curious to see if these apologies, which seemed quite frequent and apparent in their patterns, had any effect on the conversation. We implemented sentiment analysis (not a part of the transformer) to analyze the direct responses of the apologies and also used the Conversations Gone Awry Corpus conversational level metadata to see if apologies could have an effect in the overall conversational outcome.
How has this been tested?
We implemented the transformer and tested in the Conversations Gone Awry dataset. This yielded great results, as numerous examples of each type of apology were found and correctly labeled.
We did not extrapolate to any other Corpuses, however, we are curious to see how well the transformer would work. All of the rules to find and label apologies were distinctly based on patterns that we found through examination and annotation of the Corpus, so we wonder if the transformer would work well only in this setting or others as well.