Domo + Jupyter for the Window Into App Sentiment
Earlier this yr, just one of our end users made a post about how she utilized Domo and Jupyter Notebooks to run emotion detection, a form of sentiment analysis, on app evaluations from the Android App Retail outlet. We determined that we desired to share with most people some code on how this is effective. For this evaluation, we acquire opinions and characteristics from some of our preferred foodstuff supply applications, combine the info into a dataset, and build a dashboard displaying the sentiment examination of each individual review. Kendall Ruber shared her code with us, and we have recreated a dashboard down below in Domo using our Jupyter Notebook integration. Our Jupyter integration permits for far more state-of-the-art evaluation tactics and types to be designed and deployed fully within just Domo with ease. For occasion, rather of performing one thing out of the box these kinds of as a term cloud to check out and determine out what sentiment is (which can be unique from the Application Retail store Rankings), we can get a much better idea of our users’ sentiment by utilizing Python’s All-natural Language Processing strategies within Domo. We run the assessments by way of an emotion detection product, 1 Kendall located on Hugging Face, and with minimal effort and hard work we have an ML design formulated and deployed. We also can choose edge of the scheduling selections within Domo. For occasion, this Dashboard under will update day-to-day at 09:00 UTC. Other options are also offered, and present-day Domo consumers will come across the solutions common. In addition, we provide in info from the Google Engage in Keep API via python code written in Jupyter. This places data from 3rd party APIs that never nevertheless have a Domo Connector inside arrive at of your Domo atmosphere. Lastly, we also now guidance sharing and collaborating on notebooks, as nicely as integration with Accounts inside of Domo. Dashboard underneath, code and recommendations uncovered on our GitHub web page.
