Concept: Semantic Topics for Comments
Peakon's semantic topics analyzes employee comments using Natural Language Processing (NLP) to identify important themes and overall sentiment.
Unlike Concept: Keyword Topics for Comments, semantic topics groups comments into themes based on their meaning.
Considerations
- Predefined topic labels that are contextual, longer and more meaningful than keyword topics.
- Semantic topics span across multiple drivers and can form part of a wider theme.
- There's no required minimum number of comments.
- There's no required minimum number of enabled questions.
- Semantic topics generate in real time.
- Semantic topics use comments from all time.
- There's no third-party sub-processing or additional opt-ins involved. The semantic topics feature is entirely in-house.
- Although there's no required number of minimum comments to group comments under semantic topics, the dashboard must meet the confidentiality requirements set by your organization.
How it Works
How it Works
Using a combination of NLP and machine learning, Peakon groups comments with similar meaning to form topics. These comment groups display under overall themes, helping users to identify potential areas of focus.
This requires Peakon to define the topics for classification upfront instead of topic themes automatically generating, as per Concept: Keyword Topics for Comments.
It's not possible to remove topic themes from semantic topics.