Measuring the descriptiveness of web comments

This paper investigates whether Web comments are of descriptive nature, that is, whether the combined text of a set of comments is similar in topic to the commented object. If so, comments may be used in place of the respective object in all kinds of cross-media retrieval tasks. Our experiments reveal that comments on textual objects are indeed descriptive: 10 comments suffice to expect a high similarity between the comments and the commented text; 100500 comments suffice to replace the commented text in a ranking task, and to measure the contribution of the commenters beyond the commented text. Categories and Subject Descriptors: H.3.1 [Information Storage and Retrieval]: Content Analysis and Indexing—Abstracting methods; H.4.3 [Information Systems Applications]: Communications Applications—Bulletin boards General Terms: Experimentation

• Type: Conference
• Year: 2009
• Where: SIGIR
• Authors: Martin Potthast
• Keywords: Commented Text, Cross-media Retrieval Tasks, Information Retrieval, SIGIR 2009, Web Comments