|Microsoft Open Sources Natural Language Processing Tool|
|Written by Kay Ewbank|
|Monday, 29 July 2019|
Microsoft has open sourced BrowseCloud, an AI-based application that summarizes feedback data via smart word clouds, called counting grids. The application supports creating custom visualizations with your own data set and correlates metadata with topics.
Microsoft developed the tool to help them manage customer data from multiple digital channels such as survey data, email, and sites such as Reddit. The development team says that, even for internal tools teams at Microsoft, there are at least 10,000 user feedback documents generated per quarter.
BrowseCloud aims to help by summarizing feedback data via smart word clouds, called counting grids. On a word cloud, the size of the text simply scales with the frequency of the word. Text is scattered randomly on word clouds. BrowseCloud differs by taking note of the position of the word within the word cloud. To use it, a user clicks on a word in the word cloud, then scans along the visualization, seeing themes change as they move.
The application lets users add their own custom text data set, then visualize it by inspecting the largest words in clusters around the screen. You can drop a pin by clicking on the visualization to view a ranked list of verbatims (shown on the far right-hand side of the screen) related to the micro-topic you pinned.
The app also has options to search for a word to narrow down the visualization and ranked list further. You can correlate topics with positive or negative sentiment on the screen by looking at the color of the the words in a region, after applying the sentiment analysis job. You can also correlate your own custom metadata with topic.
There's a demo that you can try with a gallery of models and visualizations with data such as the Microsoft employee engagement survey, called MSPoll, and feedback on the Windows Engineering System.
The service is an ASP.NET Core Machine Learning application with Azure dependencies. Microsoft has set up Azure Pipelines for the demo site. There's also a Python command line application to train your data, and the client is a simple Angular CLI generated application.
or email your comment to: firstname.lastname@example.org
|Last Updated ( Monday, 29 July 2019 )|