With OpinionCloud we present a new opinion summarization technology for web comments in general and for YouTube and Flickr in particular. Popular web items receive up to thousands of comments; to get an idea about the crowd's overall opinion one has to read many of them, which often is impossible. The OpinionCloud summarization approach helps to retrieve this impression by generating an opinion word cloud for a given set of comments.

Opinion summarization of web comments combines two research strands: sentiment analysis and summary visualization. The former deals with the classification of words as positive, negative, or neutral, the latter deals with the design of an accessible visual representation of a set of opinions. A word's polarity can be assessed by measuring its co-occurrence with words whose polarity is known in advance. We use this idea to train a dictionary of opinion words which also contains slang terms that are typical for comments. This dictionary is used to classify the words of comments into positive, negative, and neutral. The words are arranged in a cloud where the color of a word encodes its polarity and the size of a word its frequency in the comments. The visualization is comparable to the well-known tag clouds for folksonomies.


Students: Steffen Becker