Insight into your data with new Visual Reports

We are now ready to reveal our new visualizations for analyzing text data, which are available in Gavagai Explorer! After the Gavagai Explorer reads through and structures your text data, you will immediately be taken to a dashboard with visualizations!

Topics and Sentiments are the most commonly talked about topics in the data. The height of a bar shows how often a topic is talked about, and the coloured segments inside the bar visualize the sentiment for the topic: positive, negative, or neutral.

In this image, we see the results of analyzing a set of hotel reviews. We can immediately see that the most talked about topics are about the hotel, rooms, and staff.


Sentiment Drivers are the top 4 positive and negative drivers of sentiment. Texts that include these topics have, on average, had a more positive or negative sentiment overall. This means that the factors displayed are likely to contribute to the customer’s positive or negative sentiment.

Here we see the topic Union Square is one of the most important topics for driving positive sentiment.


In the Driver Associations graphs, we look at each Driver Topic and check the Sentiment score combination for the Driver Topic and the Association to the Topic. Associations are other Topics that are mentioned and associated with the original Topic. We can then see which Associations are driving each Driver Topic.

Here we can see that when people talk about the negativity driving Topic restaurant, they often talk about it being closed and under construction.

In the Occurrence Sentiment Matrix the different topics in the data are plotted along two dimensions, sentiment and number of mentions. Topics found in the bottom right corner need to be addressed as they are mentioned frequently and in negative terms.

On the other hand, Topics in the top right corner are valued by the respondents since they are mentioned often and with a positive sentiment.


The Grade Drivers are the top 4 positive and negative drivers of the grade column in your text data. It is similar to the Sentiment Drivers graph. Texts that mention topics in this graph will, on average, have a more positive or negative effect on the Grade the customer gives you than texts that don’t mention that specific Topic. The Grade (or Net Promotor Score or similar numerical value) is found by searching from left to right the data for numerical scores.

You can always go back to your Project page and edit the Model and all your Topics by clicking Go to Project at the top of the page.


Sharing results with your colleagues is also easier than ever! Click the Publish button to get a shareable link to your dashboard.

Feel free to sign in and try it out right now. If you would like a free webinar please email sales@gavagai.io