What is sentiment analysis for the social web?
|Starts:||13:00 24 Apr 2015|
|Ends:||17:00 24 Apr 2015|
|What is it:||Workshop|
|Speaker:||Prof Mike Thelwall|
This workshop will have 2 hours based in a classroom and 2 hours in a computer suite.
Lecture (2 hours): This will describe how sentiment can be automatically detected in social web texts with human level accuracy by a multilingual program, SentiStrength, which uses a set of easily understandable, common-sense rules. SentiStrength estimates the strength of positive and negative sentiment in the short informal texts that are prevalent on the web, such as in Twitter, and is employed by Yahoo! and a range of other research-based and commercial users. The talk will give examples of applications of sentiment analysis to social media research, marketing, software product enhancement and interactive art installations but will have a particular focus on social science research applications.
Workshop (2 hours): This will involve hands-on practice with software, Mozdeh, for automatically gathering tweets matching a set of keyword searches. This session will not only use Mozdeh for gathering tweets but will also experiment with analysing it for sentiment (with SentiStrength), for topic (with word frequency analyses) and for changes over time (with time series graphs).
Travel and Contact Information
Humanities Bridgeford Street