Sentiment analysis is typically applied to unstructured data harvested online from social media, or from feedback forms etc. One leading provider of sentiment analysis is Clarabridge.
The idea is to get an indicator of the sentiment in a statement - positive, negative or neutral - by looking at the words used, or even better the combination of words used. Sentiment analysis depends on having a reference word list to compare with, where the sentiment value of particular words or phrases are listed.
I would say that rather than Sentiment Analysis helping to create more interactive dashboards, Interactive Dashboards can help you get an understanding of the bigger picture in a large amount of data onto which you apply sentiment analysis.
We have a simple, conceptual demo running on the URL below. It is powered by R in the backend and lets you search for 2 separate terms on Twitter, and make a sentiment value comparison between the hits for each search term. So for example, you can compare the use of positively charges words versus negatively charged words in tweets containing "taliban" compared to tweets containing "malala".
best regards, Theo Klemming and the Solution Team