About the appInstallation guideDocumentationReferencesSource codeLicenseFeedback guidelines
Go to about sectionHow to installUser GuideReferencesGitHubLicenseContribute


About the app

This is a PowerBI application with R script driven visual. This app attempts to help in generating topics and sentiment from text data field by writing low or no code at all. The ease of use of this application also comes with the limitation of not being scalable and sometimes needs changing some part of the code to make it relevant to the data in use.

The author has only put together a piece of code and does not own any original work, both in terms of the visualization and underlying algorithms. To this extent, the author has cited relevant researcher work in the references section.

How to install:

Usage guide:

  • If you have downloaded the demo PowerBI application shared by the author, then the app is shipped with the underlying data.
    1. In such cases, please add your own text data by using the Get Data drop down.
    2. Select the R Script visual in the canvas by clicking it. This will show the Values field in the visualizations section.
    3. Then, identify the text field from the Fields section of the app. This field is the one that needs to be analyzed for topics and sentiment.
    4. Drag and drop this text field into the Values field.
    5. Wait for the chart to populate. It generally takes less than a minute for about 300 comments with an average character length of 100.
  • If you have not downloaded the demo PowerBI application shared by the author, then please do so from the gallery


  1. Blei D.M., Ng A.Y., Jordan M.I. (2003). Latent Dirichlet Allocation. Journal of Machine Learning Research, 3, 993–1022.
  2. Phan X.H., Nguyen L.M., Horguchi S. (2008). Learning to Classify Short and Sparse Text & Web with Hidden Topics from Large-scale Data Collections. In Proceedings of the 17th International World Wide Web Conference (WWW 2008), pages 91–100, Beijing, China
  3. Lu, B., Ott, M., Cardie, C., Tsou, B.K. (2011). Multi-aspect Sentiment Analysis with Topic Models. In Proceedings of the 2011 IEEE 11th International Conference on Data Mining Workshops, pages 81–88