Tagging the blogger mood
April 23, 2006
Researchers at the Information and Language Processing Systems, Informatics Institute, University of Amsterdam, created MoodViews, tools for Blog Mood Analysis:
Our research aim is to develop novel methods for searching, discovering and retrieving blogs. We believe that non-factual aspects of blog entries such as moods are an important part of what makes people read and navigate around blogs.
At the current stage, a collection of tools is available for tracking, prediction and analysis of mood changes associated to blog posts in LiveJournal: “Bloggers using LiveJournal can report their mood at the time of writing a post; about 80% of the posts indeed have a mood attached to them.”
MoodSignals, for example, tries to explain unexpected peaks in mood-tags (there are more than 100 at a rough count) usage by looking at the lexical content of the posts (overused terms during the peak period) and news at the moment of writing (related global events). Below is a graph for the mood level for “sympathetic” around the end of August 2005; the lexical analysis reports terms as orlean, hurrican, katrina, donat, mississippi, water, prayer, devast, flood, storm and the related news all relates to Hurricane Katrina.
Folksonomy is an anarchic way of categorization: everyone can tag without fear of mistake (so, everyone increasingly does it). As the habit of tagging extends to cover more aspects of the creative process, we may profit from a bit more structure in it to make it easier (possible) to look at data from novel perspectives.